Lecture Notes: Types of Business Analytics

Introduction to Business Analytics

• Definition: Business analytics is a data management solution using tools to transform data into useful information.
  • Helps anticipate trends and make data-driven decisions.
  • Communicated through data visualization.
• Difference from Business Intelligence:
  • Business Intelligence uses historical/current data for understanding past events.
  • Business Analytics builds on BI to make predictions and decisions.

Technology and Tools in Business Analytics

• Common Tools:
  • Small organizations: Spreadsheet applications like Excel, Google Sheets.
  • Large organizations: Advanced technology such as GPUs, digital storage, high-speed networks.
• Advanced Techniques:
  • Machine learning for autonomous systems.
  • Deep learning for brain-like processing.

Workflow of Business Analytics

1. Identify the Problem/Opportunity
2. Data Collection
   • Sources: Internal (structured, e.g., databases) and External (unstructured, e.g., IoT, social media).
   • Storage: Data is pooled and centralized, usually in a data lake.
3. Data Cleaning and Storage
   • Processed through ETL into data marts and warehouses.
4. Perform Analytics: Descriptive, predictive, and prescriptive.
5. Communicate Results
   • Use presentation tools for data visualization (charts, dashboards).

Types of Business Analytics

Descriptive Analytics

• Purpose: Summarizes past events for learning and identifying patterns.
• Techniques: Data aggregation, data mining.
• Applications: Market basket analysis (e.g., beer and diapers correlation), OLAP systems (pivot tables, slicing and dicing).

Predictive Analytics

• Purpose: Predicts future outcomes based on historical data.
• Tools and Techniques:
  • Data mining, linear regression.
  • Machine learning, deep learning.
• Applications: Customer behavior analysis, fraud detection, targeted advertising.

Prescriptive Analytics

• Purpose: Advises on best actions to take for future decisions.
• Tools and Techniques:
  • Optimization, simulation, decision trees.
• Applications: Supply chain optimization, driverless car decision-making, oil and gas industry operations.

Summary

• Descriptive Analytics: Interprets past data for trends and patterns.
• Predictive Analytics: Uses statistics to forecast future.
• Prescriptive Analytics: Determines best outcomes based on scenarios.
• Choosing Method: Depends on the specific business situation.

Conclusion

• Encouragement to engage with the content (like, comment, subscribe).
• Thank you for participation.