Budapest University of Technology and Economics, Faculty of Electrical Engineering and Informatics

Name:	Position:	Department:
Csaba Gáspár-Papanek	Assistant lecturer	TMIT
Zoltán Prekopcsák	PhD student	TMIT
István Nagy	PhD student	TMIT

Basic knowledge of probability theory and statistics

The course is concerned with introducing the students to the theoretical and practical aspects of analyzing customer data. It also focuses on business practices for analytics and data mining algorithms.

The course is based on two-week cycles with three theoretical lectures and one practical laboratory. The synopsis presents the seven cycles during the course.

1. Introduction to customer analytics, presentation of the methodology and the software tools.
2. Data mining of telecommunication data
• The problem of churn, customer datasets
• Training and test sets separated in time, predicting churn, characteristic curves, profit maximization, campaign optimization
• Uplift model, segments of churning customers, rotational churn, other telecom problems
• Social network analysis and data mining
• Laboratory: Churn prediction on real telecom data
  
3. Analysis of web visitors
• Web mining and its subfields, customer behaviour, visitor identification, available data fields, basic web analytics
• Analyzing product affinity, segmenting customer groups, advantages of segmentation
• Special challenges in web mining, novel data acquisition techniques, sharing informatio
• Laboratory: E-commerce analysis on real web logs
4. Behavioural credit scoring
• Definitions of basic scorecard development
• Credit scoring based on customer behavioural data, specific issues in the management of transaction data
• Specific issues in the modelling phase: effects of seasonality, performance and sample windows
• Laboratory: developing scorecard on real customer dat 
5. Similar products, cross- and up-selling, recommender systems
• Determining similar products
• Content based and implicit recommender systems, preconditions, possible problems
• Matrix decomposition for recommender systems
• Laboratory: Comparing recommender algorithms on real data
6. Social network analysis
• Basic definitions for social network analysis, elements of the network, possibilities for building networks
• Using information from beyond the network of people
• Laboratory: social network analysis on real data
7. Additional customer analytics tasks
• Fraud detection
• Customer value calculations
• Campaign optimization
• Laboratory: Campaign optimization
• Laboratory: Customer value calculation
   

Laboratories (extracted from above):

1. Churn prediction

2. E-commerce analysis

3. Developing scorecard on real customer data

4. Recommender systems

5. Social network analysis

6. Campaign optimization

7. Customer value calculation

Lecture and laboratory

a. In the class period there is an in-class test (ZH).

b. In the examination period: homework should be written and this work should be defended at the examination (oral).

c. Condition for the signature is the pass mark of ZH test (40% above). There is a possibility to rewrite the in-class test (ZH). In the rectification period (repeat period) there is another (final) possibility to rewrite the in-class test (ZH).

d. Another condition for the signature is at least 5 attendances the laboratory exercises

There is one possibility to repeat the test in the teaching period and there is a final one in the official recap period.

On demand.

1. Michael J.A. Berry, Gordon S. Linoff: Data Mining Techniques For Marketing, Sales, and Customer Relationship Management, Wiley; 1 edition (May 27, 1997)
2. Carlo Vercellis: Business Intelligence : Data Mining and Optimization for Decision Making,  2009 John Wiley & Sons, Ltd. ISBN: 978-0-470-51138-1
3. Olivia Parr Rud:  Data Mining Cookbook: Modeling Data for Marketing, Risk and Customer Relationship Management, Wiley; 1 edition (November 3, 2000)