Course syllabus: Data Mining in Business Analytics

From Research management course
Jump to: navigation, search

The course is devoted to algorithms of intelligent data analysis and their applications. It includes numerous practical examples from the business and financial sector. Business analytics problem solutions are combined with programming tips and techniques of implementation. The course includes 14 lectures, 14 workshops, and an exam.

Course objectives:

  1. To gain an understanding of how managers use data mining to formulate and solve problems in the business and financial sectors.
  2. To learn principles of managerial decision-making support techniques.
  3. To become familiar with the algorithms needed to process, analyze and report business data.
  4. To learn how to use selected data mining software in business analytics.

Course outline:

  1. Using data mining methods in business, an introduction
    • Introduction to SciLab (workshop)
  2. Management and standards in business analytics
    • Introduction to report systems
    • Customer data collection and preprocessing
    • Data manipulation and visualization
  3. Managerial decision support
    • Delphi method, problems of voting, Pareto slicing
  4. Integral indicator construction
    • Making indicators for energy sector, ecology and finance
  5. Risk analysis and credit risk scorecards
    • Banking application scorecards developing
  6. Sociological data processing
    • Linear, ordinal and nominal scale conversion
  7. Risk model verification
    • Scorecards testing and implementation
  8. Customer multilevel modeling
    • Model comparison
  9. Customer behavior analysis
    • Clustering of telecomm customers
  10. Customer engagement and collaborative filtering
    • Nearest neighborhood and metric spaces
  11. Forecasting of goods consumption
    • Non-parametric regression
  12. Forecasting of energy consumption
    • Singular structure analysis
  13. Econometrical scenario analysis
    • Auto-regression analysis
  14. Exam and case test
    • Coursework discussion

Prerequisites: Knowledge of linear algebra and statistics is required.

Software: SciLab.

Grading: One-hour writing exam.