Michel Ballings

Learn More
The key question of this study is: How long should the length of customer event history be for customer churn prediction? While most studies in predictive churn modeling aim to improve models by data augmentation or algorithm improvement, this study focuses on a another dimension: time window optimization with respect to predictive performance. This paper(More)
We revisit well-known variables for database marketing/CRM and relationship marketing using a new methodology: Binary Bayesian Quantile regression. This method allows for a more thorough investigation of the relationship between the response variable and the covariates. The main conclusion is that taking intentions as a proxy for real churn behavior yields(More)
0957-4174/$ see front matter 2012 Elsevier Ltd. A http://dx.doi.org/10.1016/j.eswa.2012.12.007 ⇑ Corresponding author. E-mail addresses: Michel.Ballings@UGent.be (M. UGent.be (D. Van den Poel). URL: http://www.crm.ugent.be (M. Ballings). We propose an ensemble method for kernel machines. The training data is randomly split into a number of mutually(More)
In the modern Digital Era, Data Mining is the powerful area for analyzing the large data sets to get unexpected relationships (models). The analysis of statistical data on sequential data points measured at regular time interval over a period of time is time series analysis. Time series analysis is used in predicting future occurrence of a time based event.(More)
We investigate the performance of complex trading rules in equity price direction prediction, over and above continuousvalued indicators and simple technical trading rules. Ten of the most popular technical analysis indicators are included in this research. We use Random Forest ensemble classifiers using minute-by-minute stock market data. Results show that(More)
The purpose of this paper is to assess the feasibility of predicting customer churn using eye-tracking data. The eye- movements of 175 respondents were tracked when they were looking at advertisements of three mobile operators. These data are combined with data that indicate whether or not a customer has churned in the one year period following the(More)