I truly think it's insane that I can build a powerful machine learning model nowadays just by chatting to an AI assistant I remember when I first started my career doing Predictive Analytics was hard really hard with AI tools and services on the rise the barrier to learning anything really has been lowered or at the very least you can learn whatever you want quicker if you know what tools to use and how to use them in the most efficient way in today's video I'll show you the simplest way I know to build a strong machine learning model with no actual coding involved I used pcan who have kindly also sponsored this video but as always all opinions thoughts and Reflections on the tool will of course be absolutely of my own whether you're a machine learning newbie or have some or maybe lots and lots of experience training your own models I'd encourage you to stick around as you'll get to see how simple and effective it can be to utilize the latest Technologies to build build your own machine learning models regardless of your experience and I say regardless of experience because even though you can build precise models with zero coding involved you can of course adjust your code within pcan I'll show you later how so I'll walk you through how I built a revenue forecasting model but you could obviously predict other things like LTV customer turn or winback marketing mix or how to up and cross sell you can try and build your own model the same way by the way ass signing up to pcan is completely free I'll put the link in the description below along with the link to the exact data set that I used so after I logged in straight away I got to interact with the pcan AI assistant the way the assistant works is that it asked me some guiding questions to help me build my machine learning model so I answered the first question and told the AI assistant that I would like a revenue forecast model at the customer level this simply means that I want to predict what the revenue is for each and every customer the assistant is pretty smart as it recommended the subject of my prediction which was the customer of course and my target value which was the revenue and all I had to do was to confirm that all was good so far then I got another guiding question asking me how far into the future I'd like to predict the revenue for each customer you can choose whatever you want really next week month quarter or even year but I went with probably the most popular future time Horizon a month year then the tool asked me on what recurring basis I'd like to make these Revenue predictions again I went with something simplistic here so I said that I'd like the frequency of the predictions to be monthly I also had the options to trigger my predictions based on specific events for example when a customer makes a purchase but for this exercise for this model I chose to go with predicting on a monthly basis then pcan generated the Predictive Analytics question for me all I had to do was to confirm that everything still looked good and I moved on to connecting to my data I uploaded my CSV file which is actually a kaggle competition file that was cleaned up a bit and as I said before the link is in the description below so you can go ahead and download it and build your own machine learning model now of course pcon has a bunch of connectors when it comes to connecting to your data sets so if you have something sitting on SQL servers or with popular Cloud providers that's not a problem at all pcan is um capable of processing lots and lots of information as it uses data bricks so you should feel confident about loading massive data sets after the file upload was complete I simply sent it to the chat and let the tool do its work with analyzing the data the AI assistant looked at the schema of the file and recommended the column mappings of the data set based on the schema for each column type I quickly saw what the data type was I confirmed that everything looked good and moved on this time the tool actually figured it out that the user ID column was the customer ID the amount represented the revenue for each transaction and the event time was the date or timestamp that I needed again I confirmed that everything looked good and then I got a summary with my predictive question the schema my Target table Target value column Target table date column and the target entity pcan did all the heavy lifting for me in the background so now all I had to do was to click on generate notebook and then go to it to see what's actually in there think of the predictive notebook that I just generated as the brains or the control room of the entire model building process I had my SQL queries in here that the AI assistant built for me you can also see that all of the steps were clearly explained each query had a name and was actually saved kind of like a view so then the next query or sell could utilize the data from the query above for example the sampled customers table could utilize the information from the monthly sampling table so Pan's AI assistant constructed the core set for me by using the answers I provided to the guided questions now what do I mean by a coret it is the data that allowed my model to to know what it needs to learn and where it needs to learn from so the coret was a final table that consisted of all users sample dates and the target value which in my case was the revenue within 1 month after the sample date now the last thing in my notebook was the attribute section and if you have no idea what this is don't worry as I'm going to explain this in a very simple way and attribute is the data that my model will use to identify and find the patterns that will then tell my my model what the predictions will be so when the model is training it's learning the patterns to produce the correct Target values long story short use the training set to predict your Target on an ongoing basis then use your model in real life pcan also has the option to add more attributes if I wanted to so I could have easily added more by adding more data whether it was just me uploading some files or connecting to one of the many databases that are supported now of of course if I added more data I could have just used the AI assistant to replace my already existing query so that it runs on my new data set again no coding involved just pure AI magic I wanted to keep my model nice and light so I chose not to add any more attributes I just went with the one I already had my notebook had everything I needed to train my model using the data set I uploaded into the chat earlier I hit run all which ran all of my SQL queries then I trained my model now I had two options fastest which is by no surprise fast as it only takes about 10 to 30 minutes to train a model and production quality which provides better performance but takes a little longer to train I know it says several hours here but my model was actually built in about an hour or so the time it takes will of course depend on how complex your machine learning model is so once I hit train model P can quickly rans some data valid ations to check that my data was actually fit for building a predictive model and that was it I sat back and waited for about an hour and I had my Model results with some cool metrics and visualizations that helped me to gauge how good my model actually was I could see on the model evaluation tab under model performance that my model was very precise meaning the model was very close to the actual values the pan platform has a really powerful engine behind it hence the really good result results clearly their data teams know what they're doing now the model evaluation tab has a bunch of other metrics as well and I won't go through all of them but I will highlight column importance which was very useful to know in my model I could see that the amount column was by far the top contributing column to the model predictions which makes sense another helpful tab I used for interpreting my Model results was the model output tab I could quickly see what the prediction was for each user for each month I'm getting more and more used to how much and how fast AI can do things but at times it is still mindboggling to me that it allows me to do such difficult tasks like building a strong machine learning model just by interacting with an AI chat assistant now does this mean that coding is useless absolutely not the ability to read write understand and interpret code is still very very important in my opinion just think of this who can develop the better machine learning model in a shorter time frame the person who knows how to use the latest AI tools but has no real technical expertise or the person who also knows how to use the latest AI tools and has extremely strong technical skills surely the person with the strong technical skills will be able to fine-tune the model better and faster using the same AI tools compared to the person with little technical expertise not to mention all of the underlying math and statistical knowledge that is of course also essential to building good machine learning models anyway I'm going to end this video here if you enjoy content like this make sure to check out some of my other videos right here thank you so much for watching and I shall see you in the next [Music] one