hello and welcome to this new video in this video we're going to cover AI 900 real exam question and answers so before going to the question and answers I request you to kindly subscribe to our YouTube channel if you're not already a subscriber and these question and answers PDF is available to download from shaping pixel.com website the full link will be in the description so let's jump on to the AI 900 which is Microsoft azour AI fundamentals real exam questions question number one you build a machine learning model by using the Automated machine learning user interface you need to ensure that the model meets the Microsoft transparency principle for responsible AI what should you do and we have four options option A set validation type to auto option b enable explain best model option C set primary metric to accuracy and option D set max concurrent iterations to zero so the right answer here is option P enable explain best model so most businesses run on trust and being able to open the ml blackbox helps build transparency and in heav regulated Industries like healthcare and banking it is critical to compile with regulations and best practices one key aspect of these is understanding the relationship between input variables and model output knowing both the magnitude and direction of the impact each feature has own the predicted value helps better understand and explain the model with model explainability we enable you to understand feature importance as part of automated ml run runs question number two which of the following are examples of data transformation modules available in ajour machine learning designer choose two answers and we have four split data clean missing data import data data set so the right answer here is option A and option b split data clean missing data all of the options listed are examples of modules available in aour machine learning designer however only slit data and clean missing data are data transformation modules data transformation modules are used to prepare Data before a machine learning experiment common machine learning data preparation steps include cleaning missing data normalizing data and converting data from one file format to another question number three a company employs a team of customer service agents to provide telephone and email support to customers the company develops a web chat bot to provide automated answers to Common customer queries which benefits Which business benefit should the company expect as a result of creating the web chat bought solution and we have three options option A increased sales option B A reduced workload for the customer service agents and option C improved product quality reliability so the right answer here is option B A reduced workload for the customer service agents agent workload will be reduced by chatboard since it can answer simple questions question number four you are working on an application that supports an automation line in a factory several sensors provide data about equipment health and you would like to use these data in real time to identify potential line issues quickly which Azure cognitive service would be a good choice to help quickly identify issues based on your sensor data and we have four options option A anomo detector option b aure monitor option C form recognizer and option D Aur Auto Machine learning so the right answer here is option A anomal detector anomo detector is an aour cognitive service API that is used to identify anomalies in Time series data this service helps identify problems early and detect issues that might otherwise be too difficult to see without automated analysis question number five for a machine learning progress how should you split data for training and evaluation and we have four options option A use features for training and labels for evaluation option b randomly split the data into rowes for training and rows for evaluation option C use labels for training and features for evaluations and option D randomly split the data into columns for training and columns for evaluation so the right answer here is option b randomly split the data into rules for training and rules for evaluation the split data module is particularly useful when you need to separate data into training and testing sets use the split rows option if you want to divide the data into two parts you can specify the percentage of data to put in each split but by default the data is divided 50/50 you can also randomize the selection of rows in each group and use stratified sampling question number six you are using aour machine learning to develop a machine learning model to predict fuel efficiency for automobile FES manufactured between 2000 and 2010 which machine learning algorithm would be the best choice for building these model we have four options regression classification clustering reinforcement learning so the right answer here is option A regression in this scenario you are predicting a single number fuel efficiency for problems that predict a continuous value you would use a regression algorithm to build your model classification algorithms are used to predict discrete values or classes and clustering algorithms help uncover patterns and organize data into Associated clusters question number seven you are developing a model to predict events by using classification you have a confusion Matrix for the model scored on test data as shown in the following exhibit use the drop-down menus to select the answer choice that completes each statement based on the information presented in the graphic each correct selection is worth one point so there are correctly predicted positives and we have four options and there are false negatives we have four options so the right answers here is the first one it's 11 and for the second one it's 133 question number eight a customer support team would like to implement an interactive chat window on their product website that uses artificial intelligence to answer customer questions the team already has extensive frequently asked questions documentation available which aour AI solution should these team use to build a chat program from their existing frequently Asked question documentation and we have four options option A Q&A maker plus aour bot service option b Luis which is language understanding plus Azure functions option C test analytics plus Azure bot service and option D form recognizer plus aure functions so the right answer here is option A Q&A makeer plus aour bot service in this case the correct combination is to use aure Q maker as the knowledge base to power a bot using aour bot service and build a web chat bot it is also possible to use language understanding to power a bot instead of Q&A maker or in addition to Q&A maker however since the organization will use the B to answer questions from an existing frequently asked questions database Q&A maker is the correct choice question number nine for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point and the following statements are forecasting housing prices based on historical data is an example of anomo detection identifying suspicious signin by looking for deviations from usual patterns is an example of anomal detection predict whether a patient will develop diabetics based on the patient's medical history is an example of anomo detection so the right statements are the first statement is false second statement is true and the third statement is false anomo detection encompasses many important task in machine learning identifying transactions that are potentially fraudulent learning patterns that indicate that a network intrusion has occurred finding abnormal clusters of patients checking values entered into a system question number 10 to complete the sentence select the appropriate option in the answer area the the handling of unusual or missing values provided to an AI system is a consideration of the Microsoft principles for responsible Ai and we have four options inclusiveness privacy and security reliability and safety transparency so the right option here is reliability and safety AI systems need to be reliable and save in order to be trusted it is important for a system to perform as it was originally designed and for it to respond safely to new situations its inherent resident should resist intended or unintended manipulations rigorous testing and validation should be established for operating conditions to ensure that the system responds safely to edge cases an A to B testing and Champion Challenger method should be integrated into the evaluation process an AI systems performance can degrade over time so a rotus monitoring and model tracking process needs to be established to reactively and proactively measure the model's performance and retain it as necessary to modernize it question number 11 you are trained in an Ajo language understanding model and you plan to use with an interactive application that responds to user questions or commands for example a user may ask what is the temperature in Boston today what is the term used to describe what the user might say to the application in this case asking about the temperature in Boston and we have four options utterance entity intent key phrase so the right answer here is option A utterance the term utterance is used to describe the users's question or command that is the input sent to a Luis endpoint for analysis question number 12 match the types of AI workloads to the appropriate scenarios to answer drag the appropriate workload type from the column on the left to it scenarios on the right each workload type may be used once more than once not at all so each correct selection is worth one point workload types Animo detection computer vision conversational AI knowledge mining natural language processing so the right workload type for an automated chat to answer questions about refunds and exchange so the right workload is conversational AI determining whether a photo contains a person so the right answer will be computer vision determining whether a review is positive or negative which is natural language processing a computer vision image analysis service can extract a wide variety of visual features from your images for example it can determine whether an image contain adult content find specific brands or objects or find human faces natural language processing is used for tasks such as sentiment analysis topic detection language detection key phrase extraction and document categorization question number 13 you are designing an AI system that empowers everyone including people who have hearing Visual and other impairments this is an example of which Microsoft guiding principle for responsible Ai and we have four options fairness inclusiveness reliability and safety accountability so the right answer here is option b inclusiveness inclusiveness at Microsoft we firmly believe everyone should benefit from intelligent technology meaning it must incorporate and address broad range of human needs and experiences for the 1 billion people with disabilities around the world AI Technologies can be a game changer question number 14 you are working on an application that uses computer vision to identify unwanted plant species growing alongside crops in forers fields which Azure cognitive service is the best choice to help you train your image classification model and we have four options custom Vision computer vision anomo detector aour machine learning clustering analysis so the right answer here is option a custom Vision though Azure computer vision services can identify a wide range of features in images it will not work well if you need to identify domain specific features instead azour custom vision service is a no code option that you can use to train and deploy a custom image classification model question number 15 match the Microsoft guiding principles for responsible AI to the appropriate descriptions to answer track the appropriate principle from the column on the left to its description on the right each principle may be used once more than once or not at all each correct selection is worth one point so the principles are accountability fairness inclusiveness privacy and security reliability and safety so the right principle for the following statements are reliability and safety ensure that AI systems operate as they were originally designed respond to unanticipated conditions and resist harmful manipulation accountability for implementing processes to ensure that decisions made by AI systems can be overridden by humans privacy and Security provide consumers with information and controls over the collection use and storage of the data so reliability and safety to build trust it's critical that AI systems operate reliable safely and consistently under non circumstances and in unexpected conditions these systems should be able to operate as they were originally designed respond safely to anticipated conditions and resist harmful manipulations accountability the people who design and deploy AI systems must be accountable for how their systems operate organizations should draw upon industry standards to develop accountability Norms these Norms can ensure that AI systems are not the final Authority on any decisions that impacts people lives and that humans maintain meaningful control over otherwise highly account autonomous AI systems privacy and security as AI becomes more prevalent protecting privacy and securing important personal and business information is becoming more critical and complex these AI privacy and data security issues require especially close attention because access to data is essential for AI systems to make accurate and informed predictions and decisions about people AI systems must compile with the privacy laws that require transparency about the collection use and storage of data and mandate that consumers have appropriate controls to choose how their data is used question number 16 to complete the sentence select the appropriate option in the answer area when developing an AI system for self-driving cars the Microsoft Das principle for responsible AI should be applied to ensure consistent operation of the system during unexpected circumstances and we have four options so the right answer here is option C reliability and safety to build trust it's critical that AI systems operate reliable safely and consistently under normal circumstances and in unexpected conditions these systems should be able to operate as they were originally designed respond safely to unanticipated conditions and resist harmful manipulations question number 17 you plan to integrate aour Q&A maker with azour bot services to build a conversational AI system to help answer user questions what information do you need to configure a new aour P to connect with your Q&A knowledge base choose three answers and we have four options option A Q&A authorization key option b Q&A endpoint host name option C Q&A knowledge base ID and option D Q&A app service name so the right answer here is option A B and option C to configure a new Azure bot and connect with your Q&A knowledge base you will need Q&A authorization key Q&A endpoint host name Q&A knowledge base ID question number 18 your you are building an AI system which task should you include to ensure that the service meets the Microsoft transparency principle for responsible Ai and we have four options option A ensure that all visuals have an Associated text that can be ready read by a screen reader option b enable outscaling to ensure that a service scales based on demand option C provide documentation to help developers deeper code option D and show that training data set is representative of the population so the right answer here is option C provide documentation to help developers DEA code achieving transparency helps the team to understand the data and algorithms used to train the model what transformation logic was applied to the data now the final model generated and its Associated assets this information offers insights about how the model was created which allows to be reproduced in a transparent way snapshots within aour machine learning space Works support transparency by recording or returning all training related assets and metrics involved in the experiment question number 19 you have been given a data set that is unlabeled and includes detailed customer information you would like to use aju machine learning to uncover data patterns and groupings which machine learning algorithm is the best choice and we have four options clustering classification regression reinforcement learning so the right answer here is option A clustering so this is an example of a clustering machine learning problem clustering algorithms are an example of unsupervised machine learning you use clustering analysis to uncover patterns in your data and when you are working with unlabeled data you would use toring to answer a question like how is this data organized question number 20 match the types of air workloads to the appropriate scenarios to answer track the appropriate workload type from the column on the left to its scenario on the right each workload type may be used once more than once or not at all each correct celection is worth one point so workload types we have animal detection computer vision machine learning regression natural language processing so the right workload type for the statements are computer vision identify handwritten letters natural language processing we have predict the sentiment of a social media boost animal detection identify a fraudlent credit card payment machine learning regression predict next month's toy sales so question number 21 you're writing an application that analyzes customer product reviews and flags reviews that have a negative sentiment which adure cognitive s service would you use to provide sentiment analysis and Mark text as positive negative or neutral we have four options option a text analytics API option b language understanding option C speech translation and option D content moderator so the right answer here is option a text analytics API the text analytics API has the ability to take an input text document and return the predicted author sentiment positive negative or neutral question number 22 your company is exploring the use of voice recognition Technologies in its smart home devices the company wants to identify any barriers that might unintentionally leave out specific user groups these is an example of which Microsoft guiding principle for responsible Ai and we have four options accountability fairness inclusiveness privacy and security so the right answer here is option C inclusiveness so inclusiveness mandates that AI should consider all human races and experiences and inclusive design practices can help developers to understand and address potential barriers that could in unintentionally exclude people where possible speech to text text to speech and visual recognition technology should be used to empower people with hearing Visual and other impairments question number 23 what are three Microsoft guiding principles for responsible AI each correct answer represents a complete solution each correct selection is worth one point and we have the following six options knowledgeability decisiveness inclusiveness fairness option8 nness reliability and safety so the right answer here is option C D and F inclusiveness fairness reliability and safety question number 24 to to complete the sentence select the appropriate option in the answer area returning a bounding box that in indicates the location of a vehicle in an image is an example of and we have the following four options so the right answer here is object detection object detection is smaller to tagging but the API Returns the bounding box coordinates for each object found in the image for example if an image contains a dog cat and person the detect operation will list those objects with their coordinates in the image you can use this functionality to process the relationships between the objects in an image it also lets you determine whether there are multiple instances of the same object in an image question number 25 which of the following types of machine learning problems are supported with aour automated machine learning aut ml choose three answers and we have the following four options regression classification time series forecasting conversational AI so the right answer here is option A B and C aour autl supports finding the best regression classification and time series forecasting models when you configure an aut ml experiment you will need to specify which of these types of machine learning problems you are trying to solve question number 26 to complete the sentence select the appropriate option in the answer area Dash is used to generate additional features and we have the following four options so the right answer here is feature engineering feature engineering is used to generate additional features feature engineering is applied first to generate additional features and then feature selection is done to eliminate irrelevant redundant or highly correlated features question number 27 you run a charity event that involves posting photos of people wearing sunglasses on Twitter you need to ensure that you only retweet photos that meet the following requirements include one or more faces contain at least one person wearing sunglasses what should you use to analyze the images and you have the following four options option A the verify operation in the phase service option b the detect operation in the phase service option C the describe image operation in the computer vision service option D analyze image operation in the computer vision service so the right answer here is option b the detect operation in the phase service face detection is required as a first step in all the other scenarios the detect API detect human faces in an image and Returns the rectangle coordinates of their locations it also returns a unique ID that represents the stored phase data this is used in later operations to identify of verify faces optionally phas detection can extract a set of face related attributes such as head pose age emotion facial hair and glasses these attributes are General predictions not actual classifications some attributes are useful to ensure that your application is getting high quality phase data when users add themselves to phase service for example your application could advise users to take off their Sunglasses if they're wearing sunlasses question number 28 which of the following statements are true about the data used to trade train and evaluate a machine learning model and we have four options option A you should split the data set into two data sets one data set is used to train the model and the other data set is to evaluate the model option b you should always use the same data to evaluate the model that you use to train in option C the only time you need to split a data set into training and test sets is when using a regression algorithm option D when splitting a data set into training and test data sets the data should remain in order with the first 70 to 80% of the data included in the training data set and the remaining rows in the test data set so the right answers are option A you should split the data set into two data sets one data set is used to train the model and the other data set to evaluate the model so when training and evaluating a new machine learning model it is important to split the data set into a training and test data set the training data set is used to train the model and the test data set is used to evaluate the model when splitting the data the roles in the data set should be selected randomly for the split operation as a rule of thumb 70 to 80% of the data goes to the training data set with the remaining data added to the test data sets question number 29 when you design an AI system to access whether loans should be approved the factors used to make the decision should be explained this is an example of which Microsoft guiding principle for responsible Ai and we have four options transparency inclusiveness fairness privacy and security so the right answer here is option A transparency achieving transfarency helps the team to understand the data and algorithm used to train the model what transformation logic was applied to the data the final model generated and its Associated assets these information offers inside wres about how the model was created which allows it to be reproduced in a transparent way so in correct answers option b inclusiveness mandates that AI should consider all human races and experiences and inclusive design practices can help developers to understand and address potential barriers that could unintentionally exclude people where possible speech to text text to speech and visual recognition technology should be used used to empower people with hearing Visual and other impairments option C fairness is a core ethical principle that all human aim to understand and apply this principle is even more important than AI systems are being developed key checks and balances need to make sure that the systems decisions don't discriminate on run agenda race sexual orientation or religion bias towards a group or individual option D a data holder is obligated to protect the data in an AI system and privacy and security are an integral part of the system personal needs to be executed and it should be accessed in a way that doesn't compromise an individual's privacy question number 30 for each of the following statements select yes if the statement is true otherwise select no so the statements are providing an explanation of the outcome of a credit loan application is an example of the Microsoft transparency principle for responsible AI a triag bought that prioritizes insurance claims based on injuries is an example of the Microsoft reliability and safety principle for responsible AI an AI solution that is offered at different prices for different sales territories is an example of the Microsoft inclusiveness principle for responsible AI so the right answers are so the first statement is true and the second and third statements are false so the Box the first statement achieving transparency helps the team to understand the data and algorithm used to train the model what transformation logic was applied to the data the final model generated and its Associated assets these information offers insights about how the model was created which allows it to be reproduced in a transparent way the statement two a data holder is obligated to protect the data in an AI system and privacy and security are an integral part of this system personal needs to be secured and it should be accessed in a way that doesn't compromise individuals privacy and for the third statement is false inclusiveness mandates that AI should consider all human races and experiences an inclusive design practices can help developers to understand and address potential barriers that could unintentionally exclude people where possible speech to text text to speech and visual recognition technology should be used to empower people with hearing Visual and other impairments question number 31 which of the following azour cognitive Services can be integrated with azour bot Services as the engine that recognizes user intent to create an interactive chatboard application and we have four options option a language understanding option b custom Vision option C anomal detector and option D text Analytics so the right answer here is option a language understanding when you create a bot using azour bot Services you need an aure cognitive services with natural language processing to understand and respond to user input a simple solution would be to use Q&A maker that would that could be based on an existing frequenc quently asked question knowledge base however if the user interaction goes beyond a simple question and answer conversation and should include understanding and processing commands Q&A maker would not be the right choice instead the language understanding service would be the best option to integrate with the chatboard a Luis service can extract intent and entities associated with the request action question number 32 match the principles of responsible AI to appropriate requirements to answer track the appropriate principles from the column on the left to its requirements on the right each principle may be used once more than once and not at all you may need to drag the split bar between pans are scroll to view content each correct selection is worth one point so the principles we have fairness privacy and security reliability and safety transparency so the right options are for fairness the system must not discriminate based on gender race privacy and security for personal data must be visible only to approve transparency automated decisions making process must be recorded So that approved users can identify why a decision was made question number 33 you creating a bot using ajour bot service with Q&A maker as its knowledge base which of the statements regarding bot communication channels are correct and we have four options option A a web chat channel is is automatically created for you when you create a bot option b it is possible to send messages to and receive messages from a bort service using Microsoft teams option C communicating with the bort service through an email channel is not supported option D aot service can be associated with only one Communication channel so the right answer here is option A and option b when you create a bot using aour bot service a web chat channel is automatically created for you there are also many other channels that you can associate with your Bot including an email Channel and a Microsoft teams Channel a Bard service can be associated with one or more communication channels question number 34 you plan to deploy an aour machine learning model as a service that will be used by client applications which three process should you perform in sequence before you deploy the model to answer move the appropriate process from the list of process to the answer area and arrange them in the correct order so the process we have data encryption model retraining model training data preparation model evaluation the right order here is data preparation model training model evaluation to question number 35 you are building an AI based app you need to ensure that the app uses the principles for responsible AI which two principles should you follow each correct answer represents part of the solution each correct selection is worth one point and we have four options option A Implement an agile software development methodology option b Implement a process of AI model validation as a part of the software review process option C establish a risk governance committee that includes members of the legal team members of the risk management team and a privacy officer option D prevent the disclosure of the use of AI based algorithm for automated decision making so the right answer here is option b and option C question number 36 when setting up an Azure automl experiment which of the following configuration values are specified for the experiment select two answers and we have the following four options option A a primary metric used to compare the results of individual experiment runs option b a list of blocked algorithm that should be excluded from the training runs option C the host name where the best model from the experimental runs should be deployed option D the maximum time allowed to run all experimental runs so the right answer here is option A and option b so when setting up an automl experiment the primary metrix is the evaluation metrix by which each autom ml well run is compared the primary metric you choose will depend on the type of machine learning algorithm you are using regression algorithms use different primary metrics than classification algorithms also available for configuration while setting up an autom ml run is the list of blocked algorithms with this list you can explicitly specify a set of algorithms that autom ml will skip while running experiments to determine the best algorithm for a complete list of configuration settings see the automl product documentation question number 37 to complete the sentence select the appropriate option in the answer area according to Microsoft Dash principles of responsible ai ai system should not reflect biases from the data sets that that are used to train the systems the right answer here is fairness according to Microsoft fairness principle of responsibility ai ai system should not reflect biases from the data sets that I used to train the systems question number 38 match the types of AI workloads to the appropriate scenarios to answer drag the appropriate workload type from the column on the left to it scenarios on the right each workload type may be used once more than once or not at all each correct selection is worth one point so workload types we have animal detection computer vision knowledge mining natural language processing so the right workload types are knowledge Mining and an automated chatbot to answer questions about refunds and exchanges computer vision determining whether a photo contains a person natural language processing determining whether a review is positive or negative question number 39 you are using ajju's computer vision service to suggest tags that describe features identified in an image you using the computer vision rest endpoint and the Json response includes one or more tags that have been identified what other information is also included with each tag in the Json response and we have four options option A a confidence score that indicates How likely the tag correctly matches the content in the image option b b a set of coordinates given by a bounding box that identifies the tag location in the image option C a URL that provides a full description of the tag option D count indicating how many times the tag is found in the image so the right option here is option A a confidence score that indicates How likely the tag correctly matches the content in the image each tag in the response includes the name of the tag and a confidence score that indicates How likely the tag correctly matches the content in the image a confidence score is a number from 0 to one with values closer to one indicating high confidence that the tag correctly identifies features in the image tags do not include a bounding box to locate each feature if location information is needed use the object detection comp computer vision API instead question number 40 match the machine learning task to the appropriate scenarios to answer drag the appropriate task from the column on the left to its scenarios on the right each task may be used once more than once and not at all each correct selection is worth one point learning types we have feature engineering feature selection model deployment model evaluation model training so the right task for are model evaluation for examining the values of a confusion Matrix feature engineering splitting a date into month day and year Fields feature selection picking temperature and pressure to train a weather model question number 41 to complete the sentence select the appropriate option in the answer area data values that influence the prediction of a model are called and we have four options so the right option here is features question number 42 you would like to have your web chatbot application support speech output which of the following channels would you need to configure for your Bard service we have four options direct client Channel custom Channel voice Channel speech channel so the right answer here is option a direct line Channel you will need a direct line channel to support speech output for your web chatbot application so question number 43 you have the predicted versus true chart shown in the following exhibit which type of model is the chart used to evaluate and we have three options classification regression clustering so you can pause the video to have a look at the chart so the right option here is option b regression so prediction versus true shows the relationship between a predicted value and its correlating true value for a regression problem this graph can be used to measure performance of a model as the closer to the Y is equals to X line the predicted values are the better the accuracy of a predictive model question number 44 which type of machine learning should you use to predict the number of gift cards that will be sold next month and we have three options classification regression clustering so the right answer here is option be regression in the most basic sense regression refers to prediction of a numeric Target linear regression attempts to establish a linear relationship between one or more dependent variables and a numeric outcome are dependent variable you use this module to Define a linear regression method and then train a model using a label data set the trained model can then be used to make pred itions question number 45 which of the following statements best describes the characteristics of a classification model in ajur machine learning choose two answers and we have four options option A classification algorithms use labeled training data to build a model and predict the category of yet unseen data items option b classification algorithms are an example of unsupervised machine learning option C classification algorithms take unlabeled data and groups the data into two or more categories option D classification algorithms can predict both binary and multiclass classification problems so the right answer here is option A and option D classification algorithms use label training data to build a model and predict a category of unseen data items classification algorithms are an example of supervised learning and can predict either binary and multiclass classification problems it is clustering algorithms and not classification algorithms take unlabeled data and groups the data into two or more categories question number 46 you have a data set that contains information about taxi Journeys that occur during a given period you need to train a model to predict the fair of a taxi Journey what should you use as a feature and we have four options option A the number of Taxi Journeys in the data set option b the trip distance of individ idual taxi Journeys option C the fair of individual taxi Journeys option D the trip ID of individual taxi Journeys so the right answer here is option b the trip distance of individual taxi Journeys the label is the column you want to predict the identified features are the inputs you give the model to predict the label example the provided data set contains the following columns vendor ID the ID of the taxi vendor is a feature rate code the rate type of the taxi trip is a feature passenger count the number of passengers on the trip is a feature trip time in seconds the amount of time the trip took you want to predict the fair of the trip before the trip is completed at that moment you don't know how long the trip would take thus the trip time is not a feature and you will exclude this column from the model trip distance the distance of the trip is a feature payment type the payment method cash or credit card is a feature fair amount the total taxi fair paid is the label question number 47 you need to predict the C Level in meters for the next 10 years which type of machine learning should you use and we have three options classification regression clustering so the right answer here is option b regression in the most basic sense regression refers to prediction of a numeric Target linear regression attempts to establish a linear relationship between one or more dependent variables and a outcome or dependent variable use this module to Define a linear regression method and then train a model using a labeled data set the trained model can then be used to make predictions question number 48 what key piece of information do you need to call your Q&A maker service from a client application and we have four options option A the rest endpoint URL for your Q&A maker service option b the global unique Q&A maker application name option C the host name of the machine hosting your Q&A maker service and option D the region where your Q&A maker service is deployed so the right answer here is option A the rest endpoint URL for your Q&A maker service after setting up and training Q&A maker you will need the rest endpoint URL for your q& maker service question number 49 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so the statements are Automated machine learning is the process of automating the time consuming iterative tasks of machine learning model development Automated machine learning can automatically infer the training data from the use case provided Automated machine learning works by running multiple training iterations that are scored and ranked by the metrics us specific specify Automated machine learning enables you to specify a data set and will automatically understand which label to predict so the right options are so the first statement is true and the second one is false the third statement is true and the fourth statement is false Automated machine machine learning also referred to as automated ml or autom ml is the process of automating the time consuming it Rel you tasks of machine learning model development it allows data scientist analyst and developers to build ml models with high scale efficiency and productivity all while sustaining model quality statement number three is true during training AUD machine learning creates a number of pipelines in parallel to try different algorithms and parameters for you the service iterates through ml algorithms paid with feature selections where each iteration produces a model with a training score the higher the score the better the model is considered to fit your data it will stop once it hits the exit criteria defined in the experiment statement four is false apply automated machine learning when you want Jude machine learning to train and tune a model for you using the target metric you specify the label is the column you want to predict question number 50 to complete the sentence select the appropriate option in the answer area a banking system that predicts whether a loan will be repaid is an example of the dash type of machine learning and we have three options the right answer here is classification two class classification provides the answer to simple two choice questions such as yes and no are true or false question number 5 one dash occurs when a model matches training data so closely that it doesn't generalize well to other data and we have four options underfitting overfitting drift root means squared error so the right answer here is option b overfitting a common problem to be on the lookout for with any machine learning model is overfitting the data overfitting happens when the model performs very well on the training data set but does not perform well with unseen test data question number 52 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so the statements are labeling is the process of tagging training data with known values you should evaluate a model by using the same data used to train the model accuracy is always the primary metric used to measure a model's performance so the first statement is true and the second and third statements are false in machine learning if you have labeled data that means your data is marked up or annotated to show the target which is the answer you want your machine learning model to predict in general data labeling can refer to a task that include data tagging annotation classification moderation transcription or processing so the third statement is false accuracy is simply the proportion of correctly classified instances it is usually the first metric you look at when evaluating a classifier however when the test data is unbalanced are you are more interested in the performance on either one of the classes accuracy doesn't really capture the effectiveness of a [Music] classifier so question number 53 which service should you use to EXT ex ract text key value Pairs and table data automatically from scanned documents and we have four options form recognizer text analytics language understanding custom Vision so the right answer here is option a form recognizer accelerate your business processes by automating information extraction form recognizer applies Advanced machine learning to accurately extract text key value Pairs and tables from documents with just a few samples form recognizer taor its understanding to your documents both on premises and in the cloud transforms into usable data at a fraction of the time and cost so you can focus more time acting on the information rather than compiling it question number 54 Azure speech services can be used in a variety of applications which of the following correctly matches the use case with appropriate speech Service choose two answers and we have the following four options so the right answer here is option A and option b question number 55 to complete the sentence select the appropriate option in the answer area the ability to extract subtotals and totals from a receipt is a capability of the dash service and we have the following four options so the right answer here is form recognizer accelerate your business processes by automating information extraction form recognizer applies Advanced machine learning to accurately extract text key value Pairs and tables from documents with just a few samples form recognizer tailors its understanding to your documents both on premises and in the cloud turn forms into usable data at a fraction of the time and cost so you can focus more time acting on the information rather than compiling it question number 56 you use ajour machine learning designer to publish an inference pipeline which two parameters should you use to access the web service each correct answer represents part of the solution each correct selection is worth one point and we have the following four options option A the model name option b the training endpoint option C the authentication key and option D the rest end point so the right answer here is option C and option D you can consume a published pipeline in the published pipeline's page select a published Pipeline and find the rest end point of it to consume the pipeline you need the rest end point for your service the primary key for your service question number 57 which of the following features are supported by the aju text analytics API choose two answers and we have the following four options language detection named entity recognition speech to text to speech services text identification in an image so the right answers are option A and option b the aour text analytics API supports language detection and named entity recognition along with several other features language detection is used to detect the language for a given text document named entity recognition can identify people places organizations and quantities include in an input text document the other features that are listed in the choice list are available with aure cognitive Services feature user intend extraction service language understanding feature text to speech services service text to speech service feature text identification in image service computer vision Optical character recognition question number 58 to complete the sentence select the appropriate option in the answer area from ajour machine learning designer to deploy a real-time interference pipeline is a service for others to consume you must deploy the model to and we have the following four options so the right answer here is aure kuity Services AKs to perform real time inferencing you must deploy a pipeline as a realtime endpoint real time endpoints must be deployed to an aure kuties service cluster question number 59 to complete the sentence select the appropriate option in the answer area predicting how many hours of over time a delivery person will work based on the number of order received is an example of and we have the following three options classification clustering regression so the right answer here is regression in the most basic sense regression refers to prediction of a numeric Target linear regression attempts to establish a linear relationship between one or more independent variable and a numeric outcome or dependent variable you use this module to Define a linear regression method and then train a model using a labeled data set the trained model can then be used to make predictions question number 16 you have just trained and published an ajour language understanding model you plan to use the rest end point in your application to respond to user questions and commands when you send a user question or command to the rest end point for your Luis model what information is included in the response and we have the following four options a list of intents with a confidence score option b a list of potential utterances with a confidence score option C a list of entities extracted from the input question command option D the language of the input questions command so the right answer here is option A and option C the primary information included in a response from invoking a Luis endpoint with a command or question as input is the top intent a list of all intents a list of entities extracted from the input question command the confidence score assigned to each intent is a value from 0 to 1 with values closer to one indicating greater confidence that the intent was being correctly identified for more information about the Json response from invoking a Luis endpoint see the Luis document question number 61 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so the statements are aour machine learning designer provides a drag and drop visual canvas to build test and deploy machine learning models azour machine learning designer enables you to save your progress as a pipeline draft aour machine learning designer enables you to include custom JavaScript functions so the first statement is true and the second one is true and the third statement is false so the first statement is true aure machine learning designer lets you visually connect data sets and modules on an interactive canvas to create machine learning models second statement is true with the designer you can connect the models to create a pipeline draft as you edit a pipeline in the designer your progress is saved as a pipeline draft third statement is false question number 62 you have the following data set you can pause the video to have a look at the data set you plan to use the data set to train a model that will predict the house price category atories of houses what are household income and house price category to answer select the appropriate option in the answer area each correct selection is worth one point the household income we have two options a feature a label house price category we have two options a feature a label so the right options are household income a feature and house price category a label question number 63 to complete the sentence select the appropriate option in the answer Area aour machine learning designer lets you create machine learning models by and we have the following four options so the right right answer here is adding and connecting modules on a visual canvas question number 64 you're working on an application that uses aure machine learning to predict the correct medication and dosage for a patient based on their symptoms application must undergo ous testing and validation before product launch to ensure patients are given the proper medication which responsible AI practice is addressed with proper testing and validation and we have the following four options option A reliability and safety option b continuous Improvement option C accountability and option D inclusiveness so the right answer here is option A reliability and safety properly testing and validating an AI product will ensure that it performs reliably and safely question number 65 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point and we have the following three statements Automated machine learning provides you with the ability to include custom Python scripts in a training pipeline Automated machine learning implements machine learning Solutions without the need for programming experience Automated machine learning provides you with the ability to visually connect data sets and modules on an intern interactive canvas so the first statement is false the second one is true and the third statement is false question number 66 a medical research project uses a large anonymized data set of brain scan images that are characterized into predefined brain hamage types you need to use machine learning to support early detection of the different brain hamres types in the images before the images are reviewed by a person this is an example of each type of machine learning and we have the following three options clustering regression classification so the right answer here is option C classification question number 67 which azour machine learning feature offers the ability to build and deploy no code predictive models using a drag and drop interface and we have the following four options aour machine learning designer aour automated ml text analytics Q&A maker so the right answer here is option A Azure machine learning designer aour machine learning designer offers the ability to build and deploy no code predictive models using a drag and drop interface question number 68 when training a model why should you randomly split the rows into separate days of subsets and we have the following three options option A to train the model twice to attain better accuracy option b to train multiple models simultaneously to attain better performance option C to test the model by using data that was not used to train the model so the right answer here is option C to test the model by using data that was not used to train the model question number 69 you need to predict the income range of a given customer by using the following data set you can pause the video to have a look at the data set which two Fields should you use as features each correct answer represents a complete solution each correct selection is worth one point and we have the following five five options education level last name age income range first name so the right options are option A and C education level Age first name last name age and education level are features income range is a label first name and last name are Irrelevant in that they have no bearing on income age and education level are the features you should use question number 70 which of the following application features uses natural language processing NLP choose two answers and we have the following four options option A and analyze written text and highlight key phrases option b translate text from one language to another option C translate a handwritten note contained in an image into text option D locate text that is included in an image so the right answer here is option A and option b the following features use natural language processing translate text from one language to another analyze written text and highlight key phrases the other two features do not involve NLP instead these features use optical character recognition and involve identifying text within an image question number 71 you are building a tool that will process images from retail stores and identify the products of competitors the solution will use a custom model which Azure cognitive Services service should you use and we have the following four options custom Vision form recognizer phas computer vision so the right answer here is option a custom Vision question number 72 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point organizing documents into groups based on similarities of the text contained in the documents is an example of clustering grouping similar patients based on symptoms and diagnostic test results is an example of clustering predicting whether a person will develop mild moderate or severe allergy symptoms based on poent count is an example of clustering so the right options are so the first two statements are true and the third statement is false clustering is a machine learning task that is used to group instances of data into clusters that contain similar characteristics clustering can also be used to identify relationships in a data set regression is a machine learning task that is used to predict the value of the label from a set of related features question number 73 which statement about aour Automated machine learning is true choose two answers and we have the following four options option A autom ml is used to automatically select the best machine learning algorithm for your data set option b you Cano choose to use either the python SDK or a no code user interface to build an autom ml experiment option C in order to use autom ml developers must have a strong understanding of various machine learning algorithms option D an autom ml will not perform a train test split operation you must provide both a train and test data set for an autom ml experiment so the right options are option A and option b aour automl provides a no code option to select the best machine learning algorithm for your data set automatically to use autom ml users do not need to have a strong background in machine learning autom ml will automatically split your data into the appropriate train and test data sets needed to run and evaluate each experiment so question number 74 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point and you have the following four three statements a validation set includes the set of input examples that you will be used to train a mode a validation set can be used to determine find how well a model predicts labels validation set can be used to verify that all the training data was used to train the model so the right options here are so the first statement is false the second statement is true and the third statement is false the validation data set is different from the test data set that is held back from the training of the model and statement true two is true validation data set is a sample of data that is used to give an estimate of model skill while turning models hyperparameters the third statement is false the test data set not the validation set used for this the test data set is a sample of data used to provide an unbaned evaluation of a final model fit on the train in data set question number 75 what are two metrics that you can use to evaluate a regression model each correct answer represents a complete solution each correct selection is worth one point and we have the following five options coefficient of determination fub1 score root mean squared error area under curve balanced accuracy so the right answer here is option A and option C option a r s our coefficient of determination represents the predictive power of the model as a value between INF INF and 1.1.0 means there is a perfect fit and the fit can be arbitrarily poor so the scores can be negative option C RMS loss or root mean squared error measures the difference between values predicted by a model and the values observed from the environment that is being modeled so incorrect answers option b F1 score also known as balanced F score or RF measure is used to evaluate a classification model option D Au Roc or area under the curve is used to evaluate a classification model question number 76 you are using aure machine learning designer to train and evaluate a machine learning model you cannot find find the built-in model to complete a data transformation step needed to complete your workflow which of the following option is the best choice for you to complete the required data transformation and we have the following four options option A use the execute python script model with custom python code written to perform the data transformation option b use azour automl instead of azour machine learning designer option C cleanse and transform the training Data before importing into aour machine learning designer option D build your training model using the ajour machine learning apis instead of ajour machine learning designer so the right option here is option A use the execute python script model with custom python code return to perform the data transformation although aju machine learning designer has many built-in modules for building no code machine learning workflows if there is a situation where custom code is needed there are two modules you can use execute python script execute R script question number 7 7 to complete the sentence select the appropriate option in the answer area predicting how many vehicles will travel across a bridge on a given day is an example of and we have the following three options classification clustering regression so the right option here is regression regression is a machine learning task that is used to predict the value of the label from a set of related features question number 78 you need to use aour machine learning designer to build a model that will predict automobile prices which type of module should you use to complete the model to answer drag the appropriate modeles to the correct locations each module may be used once more than once or not at all you may need to drag the split bar between pans are scroll to view content each correct selection is worth one point so we have the following modules convert to CSV K means clustering linear regression split data select columns in data set summarize data so the right options are select columns in data set split data linear regression question number 79 which type of machine learning should you use to identify groups of people who have similar purchasing habits and we have the following three options classification regression clustering so the right option here is option C clustering clustering is a machine learning task that is used to group instances of data into clusters that contain similar characteristics clustering can also be used to identify relationships in a data visit question number eight when using Azure cognitive Services which of the following is not a common computer vision task and we have four options option A translate text found in an image from one language to another option b detect human faces within an image option C detect handwritten text within an image and option D classify an image based on what the image contains so the right answer here is option a while it is possible to detect text in an image translating the text from one language to another is not a computer vision task it is possible to translate text from one language to another using the translator service question number 81 to complete the sentence select the appropriate option in the answer area Dash models can be used to predict the sale price of actioned items so the right answer here is regression regression is a machine learning task that is used to predict the value of the label from a set of related features question number 82 which metric can you use to evaluate a classification model and we have four options true positive rate mean absolute error coefficient of determination root mean squared error so the right answer here is option A true person C you rate what does a good model look like an rooc curve that approaches the top left corner with 100% true positive rate and 0% false positive rate will be the best model a random model would display as a flat line from the bottom left to the top right corner wasse than random would dip below the Y is equals to X line question number 83 you work for an analytics company and are working on an application to detect credit card fraud you want to create a model to predict whether or not a particular credit card transaction is fraudulent you have historical transactions data to build your model which aju service should you use to build this model and we have four options option a aour machine learning option b anomal detection option C text analytics and option D form recognizer so the right answer here is option a aure machine learning the following services are part of azour cognitive Services form recogn izer textt analytics and AO detection cognitive Services allow developers to build AI based applications without significant AI machine learning or data science skills in a way you can consider cognitive services are machine learning as machine learning is a service this type of pre-built Services is not what is needed in the scenario presented instead you have been asked to build a custom prediction model based on your particular data set for this you would use AO machine learning there are no code options available to help build models even if you do not have significant machine learning knowledge question number 84 which two components can you drag onto a canas in azour machine learning designer each correct answer represents a complete solution each correct selection is worth one point and we have the following four options data set compute pipeline module so the right answer here is option A and option D data set and module you can drag and drop data sets and modules onto the canvas question number 85 you need to create a training data set and validation data set from an existing data set which module in the ajour machine learning designer should you use and we have the following four options select columns in data set add rows split data join data so the right answer here is option C split data a common way of evaluating a model is to divide the data into a training and test set by using split data and then validate the model on the training data use the split data module to divide a data set into two distinct sets the studio currently supports training validation data splits question number 86 which of the following azour cognitive Services would you choose to help build a conversational client application using artificial intelligence to answer customer questions and we have the following four options Q&A maker text analytics form recognizer content moderator so the right answer here is option A Q&A maker if you are if you are would like to build a bot that pulls information from existing frequently Asked question documents to answer user questions then Q&A make is the right choice to build your B's knowledge base question number 87 match the types of machine learning to the appropriate scenarios to answer track the appropriate machine learning type from the column on the left to its scenario on the right each machine learning type may be used once more than once or not at all each correct selection is worth one point so learning types we have classification clustering regression so the right learning type for the statements are regression predict how many minutes late a flight will arrive based on the amount of snowfall at an airport clustering segment customers into different groups to support a marketing department classification predict whether a student will complete a university course so statement one regression in the most basic sense regression refers to prediction of a numeric Target linear regression attempts to establish a linear relationship between monom mode independent variables and a numeric outcome or dependent variable you use this module to Define a linear regression method and then train a model using a labeled data set the train model can then be used to make predictions clustering clustering in machine learning is a method of grouping data points into similar clusters it is also called segmentation over the years many clustering algorithms have been developed almost all clustering algorithms use the features of individual items to find similar items for example you might apply clustering to find similar people by demographics you might use clustering with text analysis to group sentences with similar topics are sentiment statement three classification two class classification provides the answer to simple two choice questions such as yes no or true false question number 88 to complete the sentence select the appropriate option in the answer area Dash is the calculated probability of a correct image classification and we have the following four options so the right option here is accuracy accuracy is the calculated probability of a correct image classification question number 89 using aour machine learning designer you have just developed and deployed a machine learning model to predict a used car's price given key pieces of information about the car you are now ready to call the deployed model for predictions from an application You're Building what information do you need to make a price prediction using the deployed model choose two answers and we have the following four options option A the rest endpoint URL for the deployed model option b authentication key and option C Resource Group ID and option D inference cluster name so the right answer here is option A and option b to invoke a deployed machine learning model you will need the rest endpoint URL for the deployed model and the authentication key question number 90 to complete the sentence select the appropriate option in the answer area ensuring an AI system does not provide a prediction when important Fields contain unusual or missing values is Dash principle for responsible AI so the right answer here is reliability and safety question number 91 which statements about aure speech services capabilities are true choose two answers and we have the following four options option A the speech to text service cannot recognize or moderate profanity in an input audio file option b the speech to text API provides a batch transcript API that can be used to batch process larger audio files and option C when using the text to speech API you can configure speech settings such as speed and volume and option D speech translation is available through both the SDK and the rest API so the right answer here is option b and option C the speech to text service can recognize profanity in an input audio file and you can choose to keep it as is mask it or remove it the speech to text service offers a real time synchronous API suitable for short audio files or an A synchronized batch API that can be used to process larger input files both the text to speech and speech to text API support multiple languages when using the text to speech service you can use the speech synthesis markup language to define the synthesized speech settings for things like speaker volume and speed finally the speech translation service does not provide a rest API it is only available with a SDK question number 92 to complete the sentence select the appropriate option in the answer area ensuring that the numeric variables in training data are on a similar scale is an example of and we have the following four options so the right answer here is feature selection so ensuring that the numerical variables in training data are on similar scale is an example example of feature selection question number 93 which term completes the following definition of an aure machine learning concept A Dash is a workflow you build to manage the data and modules used to train and evaluate a machine learning model and we have the following four options pipeline experiment Jupiter notebook workspace so the right answer here is option a pipeline in ajour machine learning a pipeline is a workflow you build to manage the data and modules used to train and evaluate a machine learning model question number 94 to complete the sentence select the appropriate option in the answer area assigning classes to images before training a classification model is an example of and we have the following four options so the right answer here is labeling assigning classes to images before training a classification model is an example of labeling question number 95 which of the following types of applications are a common consumer of aure cognitive Services Q&A maker choose two answers and we have the following four options social media applications chat BS language translation applications applications used to review and moderate content text and audio so the right answer here is option A and option b Q&A maker can be used as a knowledge base when building conversational clients often associated with social media applications and chat boards question number 96 you have an ajour machine learning model that predicts product quality the model has a training data set that contains 50,000 records a sample of the data is shown in the following table you can pause the video to have a look at the table for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so the statements are mass is a feature quality test is a label temperature is a label so the right options are so the first two statements are true the third statement is false question number 97 for each of the following statement select yes if the statement is true otherwise select no each correct selection is worth one point so the statements are you train a regression model by using unlabeled data the classification technique is used to predict sequential numerical data over time grouping items by their common characteristics is an example of clustering so the right options are so the first two statements are true and the third statement is false question number 98 which of the following is not a capability of the ajour computer vision service choose two answers and we have the following four options option A describe an image with a phrase a sentence option b detect printed or handwritten text found in an image option C train a custom model to identify domain specific entities in an image option D extract text and key value pairs for from documents so the right answer here is option C and op option D of the capabilities listed there are two that are not features offered by the aour computer vision service train a custom model to identify domain specific entities is an image extract text and key value pairs from documents however these capabilities do exist with other aure cognitive services for example it is possible to train and publish a custom model to identify and locate domain specific entities in an image using aure custom Vision Services also you can use form recognizer to extract text and key value pairs from documents so question number 99 which two actions are performed during the data inje and data preparation stage of an ajour machine learning process each correct answer represents part of the solution each correct selection is worth one point and we have the following five options option A calculate the accuracy of the model option b score test data by using the model option C combine multiple data sets option D use the model for realtime predictions and option e remove records that missing values so the right answer here is option C and option e question number 100 you need to predict the animal population of an area which Azure machine learning type should you use and we we have the following three options regression clustering classification so the right answer here is option A regression regression is a supervised machine learning technique used to predict numeric values question number 101 which of the following are one of the Aur cognitive Services six principles of responsible AI choose two answers and we have the following four options transparency fairness communication collaboration so the right answer here is option A and option b transparency and fairness the J cognitive Services six principles of responsible AI are fairness transparency inclusiveness accountability safety and reliability security and privacy question number 102 which two languages can you use to write custom code for aour machine learning design each correct answer represents a complete solution each correct selection is worth one point we have the following four options python r c Scala so the right answer here is option A and option b Python and R language use Azure machine learning designer for customizing using Python and R code question number 103 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point and we have the following three statements for a regression model labels must be numeric for a clustering model labels must be used for a classification model labels must be numeric so the right options are the first statement is true and the second and third statements are false question number 104 you are building an application that allows you to search the transcripts of recorded conversations from your customer support call center which Azure cognitive service would you use to convert audio from call center conversations into text and we have the following four options speech to text language understanding translator text analytics so the right answer here is option a speech to text aour speech to text service can convert audio files to text and supports a variety of languages question number 105 your company wants to build a recycling machine for bottles the recycling machine must automatically identify bottles of the correct shape and reject all other items which type of AI workload should the company use and we have the following four options Animo detection conversational AI computer vision natural language processing so the right option here is option C computer vision aour computer vision service gives you access to Advanced algorithms that process images and return information based on the visual features you are interested in for example computer vision can determine whether an image contain adult content find specific brands or objects or find human faces question number 106 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point statements are when creating an object detection model in the custom vision service you must choose a classification type of either multi-label or multic class you can create an object detection model in the custom vision service to find the location of content within an image so when creating an object detection model in the custom vision service you can select from a set of predefined domains so the right options are so the first statement is false and the second and third statements are true so question number 107 in Aur cognitive Services what is a Q&A knowledge base and we have the following four options option A a data set that includes question and answer pays option b a data set that includes a list of erences and intents option C a data set that includes a set of Industry specific terms and the corresponding definition option D a database that includes training data sets for a variety of conversational AI scenarios so the right answer here is option A a data set that includes question and answer pairs when you set up Q&A maker you create a a knowledge base that includes a set of questions and answer pairs each answer can have multiple forms of the same question once your Q&A maker service is deployed it can take a user questions as input and process it to determine the appropriate answers found in the knowledge base question number 108 in which two scenarios can you use the form recognizer service each correct answer represents a complete solution each correct selection is worth one point and we have the following four options option A extract the invoice number from an invoice option b translate a form from French to English option C find image of product in a catalog option D identify the retailer from a receipt so the right answer here is option A and option [Music] D question number 109 select the answer that correctly completes the sentence counting the number of animals in an area based on a video feed is an example of and we have have the following four options so the right answer here is computer vision counting the number of animals in an area based on a video feed is an example of computer vision question number 110 you want to build a web based application that uses conversational AI to answer a variety of user questions which aju service is the best choice and we have the following four options Aur bot service translator text analytics speech translation so the right answer here is option A Aur bot service the aure bot service provides the bot framework SDK and Bot framework composer to help teams build and deploy Bots a board is a service capable of interacting with the user in a similar way to how the user would interact with another person the purpose of a bot is to automate relatively simple and repetitive task so that human intervention is not needed you have a database that contains a list of employees and their photos you're tagging new photos of the employees for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point you have the following three statements the phase service can be used to perform facial recognition for employees the phase service will be more accurate if you provide more sample photos of each employees from different angles if an employee is wearing sunglasses the face service will always fail to recognize the employee so the right options are so the first two statements are true and the third statement is false question number 112 you need to develop a mobile app for employees to scan and store their expenses while traveling which type of computer vision should you use and we have the following four options semantic segmentation image classification object detection optical character recognition so the right answer here is option D optical character recognition Aus computer vision API includes optical character recognition capabilities that extract printed or handwritten text from images you can extract text from images such as photos of license plates or containers with serial numbers as well as from documents invoice bills Financial reports articles and more question number 113 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so we have the following statements the custom vision service can be used to detect objects in an image the custom vision service requires that you provide your own data to train the model the custom vision service can be used to analyze video files so the right answers are so the first two statements are true and the third statement is false so the first statement is true custom Vision functionality can be divided into two features image classification applies one or more labels to an image object detection is similar but it also Returns the coordinates in the image where applied labels can be found and option two is true the custom vision service uses a machine learning algorithm to analyze images you the developer submit groups of images that feature the lack of the characteristics in question you label the images yourself at the time of submission then the algorithm trains to these data and calculates its own accuracy by testing itself on those same images the third statement is false custom vision service can be used only on graphic files question number 114 you are processing photos of runners in arrays you need to read the numbers on the runnner chat to identify the runners in the photo which type of computer vision should you use and we have the following four options facial recognition optical character recognition image classification object detection so the right option here is option b optical character recognition optical character recognition allows you to extract printed or handwritten text from images and docum doents question number 115 match the types of machine learning to the appropriate scenarios to answer track the appropriate machine learning type from the column on the left to its scenarios on the right each machine learning type may be used once more than once or not at all each correct selection is worth one point so machine learning types we have facial detection facial recognition image classification object detection optical character recognition semantic segmentation so the right machine learning types are image classification separate image of polar bear and brown bear object detection determine the location of a bear in a photo semantic segmentation determines which pixels in an image are part of of a beer so image classification is a supervised learning problem Define a set of Target classes and train a model to recognize them using labeled example photos an object detection is a computer vision problem while closely related to image classification object detection performs image classification at a more glandular scale object detection both locates and categorizes entities within images and semantic segmentation achieves fine grained inference by making dense predictions inferring labels for every pixel so that each pixel is labeled with a class of its enclosing objects or origin question number 116 use drones to identify where weeds grown between rows of crops to send an instructions for the removal of the weeds this is an example of each type of computer vision and we have the following three options object detection optical character recognition seen segmentation so the right option here is option A op object detection object detection is similar to tagging but the API Returns the bounding box coordinates for each tag applied for example if an image contains a dog cat and person the detect operation will list those objects together with their coordinates in the image so the incorrect options are optical character recognition allows you to extract printed or hand wred handwritten text from images and documents scene segmentation determines when a scene changes in a video based on visual cues a scene depicts a single event and it's composed by a series of consecutive shs which are semantically related question number 217 Ma the facial recognition task to the appropriate questions to answer drag the appropriate task from the column on the left to its question on the right each task may be used once more than once not at all each correct selection is worth one point so we have the following task grouping identification similarity verification so the right task for the statements are verification doe two images of a phase belong to the same person similarity does this person look like other people grouping do all the faces Belong Together identification who this person in this group of people so phase verification checks the likelihood that two faces belong to the same person and receives a confidence score and phase detection detect one or more human faces along with attributes such as age emotion pose smile and facial hair including 27 landmark of each face in the image question number8 match the types of computer vision workloads to the appropriate scenarios to answer drag the appropriate workload type from the column on the left to its scenarios on the right each work workload type may be used once more than once or not at all each correct selection is worth one point so the workload types we have facial recognition image classification object detection optical character recognition OCR so the correct workload type for the statements are facial recognition you identify celebrities in an image optical character recognition extract movie title names from movie poster image object detection we have locate vehicles in images question number 119 you need to determine the location of cars in an image so that you can estimate the distance between the cars which type of computer vision should you use and we have the following four options optical character recognition object detection image classification face detection so the right answer here is option b object detection object detection is similar to tagging but the API Returns the bounding box coordinates for each object found for example if an image contains a do c and person the detect operation will list those objects together with their coordinates in the image you can use this function it to process the relationships between the objects in an image it also lets you determine whether there are multiple instances of the same tag in an image the detect API applies tags based on the objects are living things identified in the image there is currently no formal relationship between the tagging taxonomy and the object detection taxonomy at a conceptual level the detect API only finds objects and living things while the tag API can also include contextual terms like indure which can't be localized with bonding boxes question number 120 to complete the sentence select the appropriate option in the answer area you can use the dash service to train an object detection model by using your own images and we have the following four options computer vision custom Vision form recognizer video indexer so the right option here is custom Vision hajur custom vision is a cognitive service that lets you build deploy and improve your own image classifiers an image classifier is an EI service that applies labels to images according to their visual characteristics unlike the computer vision service custom Vision allows you to specify the labels to apply the custom vision service uses a machine learning algorithm to apply labels to images you the developer must submit groups of images that feature and lack the characteristics in question you label the images yourself at the time of submission then the algorithm trains to these data and calculates its own accuracy by testing itself on on those same images once the algorithm is trained you can test retrain and eventually use it to classify new images according to the needs of your app you can also export the model itself for offline use so question number 121 you send an image to a computer vision API and receives back the unnoted annotated image shown in the exhibit which type of computer vision was used and we have the following four options object detection phas detection optical character recognition image classification so the right option here is option a object detection object detection is a similar to tagging but the API Returns the founding box coordinates in for each object found for example if an image contains a dog cat and a person the detect operation will list those objects together with their coordinates in the image you can use this functionality to process the relationships between the objects in an image it also lets you determine whether there are multiple instances of the same tag in an image the detect API applies tags based on the objects are living things identified in the image that is currently no formal relationship between the tagging taxonomy and object detection taxonomy at a conceptual level the detect API only finds objects and living things while the tag API can also include contextual terms like indoor which can't be localized with bounding boxes question number 122 what are two two tasks that can be performed by using the computer vision service each correct answer represent presents a complete solution each correct selection is worth one point and we have the following four options option A train a custom image classification model option b detect faces in an image option C recognize handwritten text option D translate the text in an image between languages so the right option here is option b and option C so Azure computer vision service provides developers with access to Advanced algorithms that process images and return information based on the visual features you are interested in for example computer vision can determine whether an image contains adult content find specific brands or objects or find human faces computer vision includes optical character recognition capabilities you can use the new read API to extract printed and handwritten text from images and documents question number 123 what is a use case for classification and we have the following four options option A predicting how many cups of copy a person will drink based on how many hours the person slept the previous night option b analyzing the contents of images and grouping images that are have similar colors option C predicting whether someone uses a bicycle to travel to work based on the distance from home to work option D predicting how many minutes it will take someone to run a race based on past race times so the right option here is option C predicting whether someone uses a bicycle to travel to work based on the distance from home to work two class classification provides the answer to simple two choice questions such as yes or no or true or false so incorrect answers are this is regression this is clustering and option D is regression so question number 124 what are two tasks that can be performed by using computer vision each correct answer represents a complete solution each correct selection is worth one point and we have the following five options predict stock prices detect brands in an image detect the color scheme in an image translate text between languages extract key phrases so the right option here is option b and option C identify commercial brands in images or videos from a database of thousands of global logos you can use this feature for example to discover which brands are most popular on social media are most prevalent in media product placement option C analyze color usage within an image computer vision can determine whether an image is black or white or color and for color images identify the dominant and accent colors question number 125 you need to build an image taging solution for social media that tags images of your friends automatically which aure cognitive Services service should you use and we have the following four options phase form recognizer text analytics computer vision so the right option here is option A phas the aure cognitive Services pH service provides facial recognition and Analysis capabilities it can detect and recognize faces in images identify specific individuals and analyze facial attributes such as age gender emotions and more by trading the phase service with images of your friends you can create a custom phas recognition model that will allow you to automatically tag images of your friends in social media the face service this provides rust and accurate phas detection and recognize capabilities making it suitable for tasks like image tagging based on specific individuals question number 126 in which two scenarios can you use the form recognizer service each correct answer represents a complete solution each correct selection is worth one point and we have the following four options option A identify the retailer from a receipt option b translate from French to English option C extract the invoice number from an invoice option D find images of products in a catalog so the right answer here is option A and option C question number 127 drag drop match the facial recognition task to the appropriate questions to answer drag the appropriate task from the column on the left to its question on the right each task may be used once more than once or not at all each correct selection is worth one point so we have the following task grouping identification similarity and verification so let's see the right task for the statements now verification do two images of a face belong to the same person similarity does this person look like other people identification who is this person in this group gr of people so the first one verification identity verification so modern Enterprises and apps can use a phase identification and phase verification operations to verify that a user is who they claim to be and similarity the find similar operation does face matching between a Target phase and a set of candidate phases finding a small small a set of faces that look similar to the Target phase this is useful for doing a phase search by image the service supports two working modes match person and match face the match person mode returns similar faces after filtering for the same person by using the verif verify API the match phase mode ignores the same person filter it returns a list of similar candidates faces that may or may not belong to the same person and identification phase identification can address one to many matching of one phase in an image to a set of Faces in a secure repository match candidates are returned based on how closely their face data matches the query phase this scenario is used in granting building or airport access to a certain group of people are verifying the user of a device question number 128 which computer vision feature can you use to generate automatic captions for digital photographs and we have the following four options recognize text identify the areas of Interest detect objects describe the images so the right options here is option D describe the images describe images with human readable language computer vision can analyze an image and generate a human readable phrase that describes its contents the algorithm return returns several descriptions based on different visual features and each desp description is given a confidence score the final output is a list of descriptions ordered from highest to lowest confidence the image description feature is part of the analyze image API question number 129 which service should you use to extract text key value pass and table data automatically from scan documents and we have the following four options custom vision face form recognizer language so the right option here is option C form recognizer form recognizer applies Advanced machine learning to accurately extract text key value pairs tables and structures from documents so question number 130 select the answer that correctly completes the sentence Dash extracts text from handwritten documents and we have the following four options so the right option here is optical character recognition OCR extract text from handwritten documents handwritten OCR optical character recognition is the process of automatically extracting handwritten information from paper scans and other low quality digital documents question number 131 on you are developing a solution that uses the text analytics service you need to identify the main talking points in a collection of documents which type of natural language processing should you use and we have the following four options entity recognition key phrase attraction extraction sorry sentiment analysis language detection so the right option here is option b key phrase extraction broad entity extraction identify important Concepts in text including key key phrase extraction broad entity extraction identify important Concepts in text including key phrases and named entities such as people faces and organizations question number 132 in which two scenarios can you use speech recognition each correct answer represents a complete solution each correct selection is worth one point and we have the following four options option A any incar system that reads text messages allowed option b providing closed captions for recorded or live videos option C creating an automated public address system for a train station option D creating a transcript of a telephone call or meeting so the right option here is option b and option D question number 133 to complete the sentence select the appropriate option in the answer area while presenting at a conference your session is transcribed into subtitles for the audience this is an example of and we have the following four options sentiment analysis speech recognition speech synthesis translation so the right option here is speech recognition question number 134 you need to build an app that will read receip instructions allowed to support users who have reduced Vision which version service should you use and we have the following four options text analytics translator speech language understanding so the right option here is option C speech question number 135 for each of the following statement select yes if the statement is true otherwise select no each correct selection is worth one point so the we have the following statements you can use the speech service to transcribe a call to text you can use the text andex service to extract key entities from a call transcript you can use the speech service to translate the audio of a call to a different language so the right options are so all the statements are true so question number 136 your website has a chart bot to assist customers you need to detect when a customer is upset based on what the customer types in the chatboard which type of AI workload should you use we have the following four options Animo detection computer vision regression name natural language processing so the right option here is option D NLP natural language processing natural language processing NLP is used for task such as sentiment analysis topic detection language detection key phrase extraction and document categorization sentiment analysis is the process of determining whether a piece of writing is positive negative or neutral question number 137 you plan to develop a bot that will enable users to carry a knowledge Base by using natural language processing which two Services should you include in the solution each correct answer represents part of the solution each correct selection is worth one point we have the following four options Q&A maker Aur bot service form recognizer anomo detector so the right answer here is option A and option b Q&A maker and Aur B service question number 138 in which two scenarios can you use a speech synthesis solution each correct answer represents a complete solution each correct selection is worth one point and we have the following four options option A an automated voice that reads back a credit card number entered into a telephone by using a numeric keypad option b generating live captions for a news broadcast option C extracting key phrases from the AUD recording of a meeting option D an AI character in a computer game that speaks audibly to a player so the right option here is option A and option b AO text to speech is a speech service feature that converts text to life like speech so question number 139 for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point so you have the following three statements you can use the translator service to translate text between languages you can use the translator service to detect the language of given text you can use the translator service to transcribe audible speech into text so the right options are so the first two statements are true and the third statement is false the translators service provides multi- language support for text translation transliteration language detection and dictionaries speech to text also known as automatic speech recognition is a feature of speech services that provides transcription question number 140 you need to scan the news for articles about your customers and alert employees when there is a negative article positive articles must be added to a press book which NLP task should you use to complete the process to answer drag the appropriate task to the correct locations each task may be used once more than once not at all you may need to track the split bar between pans or scroll to view content so we have the following task entity recognition sentiment analysis speech synthesis and translation so we have the following flow so let's see the light right tasks in the boxes are so entity recognition in the first one and sentiment analysis in the second one question number 141 you're building a knowledge Base by using Q&A maker which which file format can you use to populate the knowledge base and we have the following four options ppdx XML zp and PDF so the right option here is option D PDF so content types of documents can add to knowledge base content types include many standard structured documents such as PDF Doc and txt the tool supports the following file formats for inje tsv a Q&A contained in the format question answer txt do dox and.pdf Q&A contained as regular frequently Asked question content that is a sequence of question and answers so the incorrect answers are pptx is a default presentation file format for a new PowerPoint presentations and option b it is not possible to injest XML file directly so question number 142 in which scenario should you use key phrase extraction and we have the following four options option A identifying whether reviews of a restaurants are positive or negative option b generating captions for a video based on the audio track option C identifying Which documents provide information about the same topics option D translating a set of documents from English to German so the right option here is option C identifying which documents provide information about the same topics so question number 143 you have Insurance claim reports that are stored as text you need to extract key terms from the reports to generate summaries which type of AI workload should you use and we have the following four options NLP natural language processing conversational AI Animo detection computer vision so the right option here is option A NLP natural language processing question number 144 to complete the sentence select the appropriate option in the answer area natural language processing can be used to and we have the following four options so the right option here is classify email messages as work related or personal so NLP natural language processing is used to task such as sentiment analysis topic detection language detection keyphrase extraction and document categorization so question number 1445 which AI service can you use to interpret the meaning of a user input such as call me back later and we have the following four options translator text analytics speech language understanding so the right option here is option D language understanding so language understanding is a cloud-based AI service that applies custom machine learning intelligence to users conversation natural language text to predict overall meaning and pull out relevant detailed information question number 146 you are developing a chatbot solution in aour which service should you use to determine a user's intent and we have the following four options translator Q&A maker speech language understanding so the right option here is option D language understanding so language understanding is cloud-based API service that applies custom machine learning intelligence to a user conversational natural language text to predict overall meaning and pull out relevant detailed information design your Luis model with categories of user intentions called intents each intent needs example of user utterances each utterance can provide data that needs to be extracted with machine learning entities question number 147 you need to make that written press release for a company available in a range of languages which service should you use we have the following four options translator text analytics speech language understanding so the right option here is option a translator translator is a cloud-based machine translation service you can use to translate text in near real time through a simple rest API call the service uses modern neural machine translation technology and offers statistical machine translation technology custom translator is an extension of translator which allows you to build neural translation systems so question number 148 for each of the following statements select yes if the statement is true otherwise I select no so each correct selection is worth one point so we have the following three statements the text analytics service can identify in which language text is written the text analytics service can detect handwritten signatures in a document the text analytics service can identify companies and organizations mentioned in a document so the right options are so the first statement is true the second statement m is false and the third statement is true the text analytics API is a cloud-based service that provides Advance natural language processing over raw text and includes four main functions sentiment analysis key phrase extraction named entity recognition and language detection so the first statement is true you can detect which Lang language that input text is return return in and report a single language quote for every document submitted on the request in a wide range of languages variants dialects and some Regional cultural languages the language code is paired with a score indicating the strength of the score so this third statement is true named entity recognition identify the categorize entities in your text as people places organizations date time quantity percentages currencies and more well-known entities are also recognized and linked to more information on the web so question number 149 match the types of natural languages processing workloads to the appropriate scenarios to answer drag the appropriate workload type from the column on the left to its scenarios on the right each workload type may be used once more than once or not at all each correct selection is worth one point so workload types we have entity recognition key phrase extraction language modeling sentiment analysis translation speech recognition and speech synthesis so the right of clo types for the statements are entity recognition for extract persons locations and organizations from the text and sentiment analysis for evaluates text along a positive negative scale translation for converts text to a different language for each of the following statements select yes if the statement is true otherwise select no each correct selection is worth one point and we have the following three statements monitoring online service reviews for profanities is an example of natural language processing the second statement identifying brand logos in an image is an example of natural language processing third statement monitoring public new sites for negative mentions of a product is an example of natural language processing so the right options are so the first statement is true the second statement is false and third statement is true so question number 151 you are developing a natural language processing solution in ajour the solution will analyze customers reviews and determine how positive and negative each review is this is an example of which type of natural language process procing workload and we have the following four options language detection sentiment analysis key phrase extraction entity recognition so the right option here is option b sentiment analysis sentiment analysis is the process of determining find ing whether a piece of writing is positive negative or neutral so question number 152 you use natural language processing to process text from a Microsoft new story you receive the output shown in the following exhibit you can pause the video to have a look at the exhibit so which type of natural language processing was performed and we have the following four options entity recognition key phrase extraction sentiment analysis translation so the right option here is option A entity recognition so named entity recognition is the ability to identify different entities in text and categorize them into predefined classes or types such as person location event protu and organization in this question the square bracket indicates the entity such as date time person type skill