Academic Papers about Predictive Analytics

This website bundles scientific research about predictive analytics. The customer intelligence cluster of the Department of Marketing at Ghent University tries to be at the forefront of new developments in this field.

Predictive analytics is trying to find structure in historical data to come up with predictions about future events by using a variety of techniques from data mining, machine learning, and statistics. In this process, one usually constructs models to capture (structural) relationships among factors that potentially have an impact on future events.

Here you will find direct links to the full papers available for download. Publishers forced us to take down the full text of some of the papers. Please contact dirk dot vandenpoel AT ugent dot be for a copy in that case.

Please use the navigation menu on the left <------ to visit our different research topics.

Marketing Analytics / Predictive Analytics for Marketing

Two key ingredients lead to good predictions: 1. Good data, 2. Good models.

Our experience in doing hundreds of projects has taught us that the former is actually more important than the latter. So we organized our contributions in this field according to the type of data being used in the project:

VAN DEN POEL Dirk, Predicting Mail-Order Repeat Buying: Which Variables Matter?, Tijdschrift voor Economie & Management, 48 (3), 371-403.

using situational factors

BAECKE Ph & VAN DEN POEL D. (2010), Improving purchasing behavior predictions by data augmentation with situational variables, Forthcoming in International Journal of Information Techonology and Decision Making.

using text

THORLEUCHTER D., VAN DEN POEL D., PRINZIE A. (2010), Mining Ideas from Textual Information, Expert Systems with Applications, Forthcoming

COUSSEMENT K., VAN DEN POEL D. (2009), Improving customer attrition prediction by integrating emotions from client/company interaction emails and evaluating multiple classifiers , Expert Systems with Applications, 36 (3), 6127-6134.

COUSSEMENT K., VAN DEN POEL D. (2008), Integrating the Voice of Customers through Call Center Emails into a Decision Support System for Churn Prediction, Information and Management, 45 (3), 164-174.

COUSSEMENT K., VAN DEN POEL D. (2008), Improving Customer Complaint Management by Automatic Email Classification Using Linguistic Style Features as Predictors, Decision Support Systems, 44 (4), 870-882.

using sequential information

PRINZIE Anita & VAN DEN POEL Dirk (2007), Predicting home-appliance acquisition sequences: Markov/MTD/MTDg and survival analysis for modeling sequential information in NPTB models, Decision Support Systems, 44 (1), 28-45.

PRINZIE Anita & VAN DEN POEL Dirk (2006), Investigating Purchasing Patterns for Financial Services using Markov, MTD and MTDg Models, European Journal of Operational Research, 170 (3), 710-734.

using clickstream data

DE BOCK K.W. & VAN DEN POEL D. (2010), Predicting website audience demographics for web advertising targeting using multi-website clickstream data, Fundamenta Informaticae, 98 (1), 49-67.

using psychographics

BAECKE Ph & VAN DEN POEL D. (2010), Data Augmentation by Predicting Spending Pleasure Using Commercially Available External Data, Forthcoming in Journal of Intelligent Information Systems.

Predictive Analytics for Credit Scoring

Using Ensembles for Predictive Analytics

If you extracted the last bit of information out of your data, then the last way to make a difference may be to move to more sophisticated models. Lately, ensemble methods have taken over the lead in predictive performance.

PRINZIE Anita & VAN DEN POEL Dirk (2008), Random Forests for Multiclass classification: Random Multinomial Logit, Expert Systems with Applications, 34 (3), 1721-1732.

DE BOCK K.W., COUSSEMENT K., VAN DEN POEL D. (2010), Ensemble classification based on generalized additive models , Computational Statistics and Data Analysis, 54 (6), 1535-1546.

Education / Training

At the Department of Marketing we also have a specialized Master program in predictive analytics: "Master of Marketing Analysis" (a one-year master program), as well as "Marketing Engineering" (a two-year master as part of Business Engineering).