Developing Churn Models Using Data Mining Techniques and Social Network Analysis

Developing Churn Models Using Data Mining Techniques and Social Network Analysis

Goran Klepac (Raiffeisenbank Austria Zagreb, Croatia), Robert Kopal (University College for Law and Finance Effectus Zagreb, Croatia & University College for Applied Computer Engineering Algebra Zagreb, Croatia) and Leo Mršić (University College for Law and Finance Effectus Zagreb, Croatia & University College for Applied Computer Engineering Algebra Zagreb, Croatia)
Indexed In: SCOPUS
Release Date: July, 2014|Copyright: © 2015 |Pages: 308
ISBN13: 9781466662889|ISBN10: 1466662883|EISBN13: 9781466662896|DOI: 10.4018/978-1-4666-6288-9


Churn prediction, recognition, and mitigation have become essential topics in various industries. As a means for forecasting and managing risk, further research in this field can greatly assist companies in making informed decisions based on future possible scenarios.

Developing Churn Models Using Data Mining Techniques and Social Network Analysis provides an in-depth analysis of attrition modeling relevant to business planning and management. Through its insightful and detailed explanation of best practices, tools, and theory surrounding churn prediction and the integration of analytics tools, this publication is especially relevant to managers, data specialists, business analysts, academicians, and upper-level students.

Topics Covered

The many academic areas covered in this publication include, but are not limited to:

  • Attribute Relevance Analysis
  • Behavioral Variables
  • Churn Prediction
  • Data Sampling
  • Predictive Analytics
  • Risk Management
  • Web Analytics

Reviews and Testimonials

Churn prediction is a data mining technique that predicts if and when a current customer will stop using a particular service or will change to a different organization to fulfill their needs. Such models are invaluable in managing risk and eliminating customer attrition. In this book, the authors attempt to combine the numerical accuracy with the value judgments inherent in social network analysis to provide more predictive churn models for online businesses and organizations. The book uses several case models to show how to create and plan a churn modeling project, how to maintain it to provide accurate and relevant data, and how to analyze the results of their churn models to reduce customer attrition. Although intended for mangers, business analysts and data specialists, the book also provides relevant information for use in an upper level business class.

– ProtoView Book Abstracts (formerly Book News, Inc.)

Table of Contents and List of Contributors

Search this Book:

Author(s)/Editor(s) Biography

Goran Klepac, PhD, University College Professor works as a head of Strategic unit in Sector of credit risk in Raiffeisenbank Austria Inc, Croatia, Europe. In several universities in Croatia, he lectures subjects in domain of data mining, predictive analytics, decision support system, banking risk, risk evaluation models, expert system, database marketing and business intelligence. As a team leader, he successfully finished many data mining projects in different domains like retail, finance, insurance, hospitality, telecommunications, and productions. He is an author/coauthor of several books published in Croatian and English in domain of data mining.

Robert Kopal, PhD, Dean of University College for Law and Finance Effectus in Zagreb, Croatia, Europe; University College Professor; lecturer at several university colleges in Croatia, and at CROMA (Croatian Managers' and Entrepreneurs' Association) EduCare Program; author & co-author of six books (on competitive intelligence analysis, game theory, etc.), numerous chapters in books of various authors, and more than forty scientific and professional papers; workshop manager and teacher at more than hundred business and intelligence analysis workshops; designer of several specialized IT systems; certified trainer in the area of structured intelligence analysis techniques and SW; SCIP and IALEIA member; held presentations at various national and international conferences; participated in and led a number of national and international intelligence analysis projects.
Leo Mršic, PhD, University College Professor graduated with a major in insurance; earned MSc degree with a major in business statistics; earned PhD degree in data science all at University of Zagreb, Croatia, Europe. Combining business and technology approach, he has great field experience related to many industries like retail, insurance, finance, ICT, business law and project management. Has relevant top management and consulting experience participating in many projects across the supply chain with focus on retail. Active in conferences and guest lecturer on several university programs related to various aspects of business like management, consumer behavior, data science/data mining and managing business risks. Co-author on several books in area of data science and it’s appliance in business. Member of the board at Croatian Oracle Users Group, member and mentor at Young Executives Society in Croatia, member of Croatian Association of Court Expert Witness in Croatia.