The VUB Data Analytics lab is headed by prof. Wouter Verbeke and prof. Tias Guns, at the Faculty of Economics and Social Sciences and Solvay Business School. The lab has strong ties to the MOBI research group and collaborates with several academic and industrial partners.
Our focus is on developing and adapting analytical tools and algorithms to take into account the operational setting, domain expertise and business requirements of the application at hand. This differs from the typical statistical or IT perspective, and takes into account the operational context; including impact on revenues, adherence to user requirements, interpretability and interaction with other systems.
Example application domains include demand forecasting, exploratory data analysis, credit risk modeling, constraint-based clustering, fraud detection and customer churn prediction.
Prof. Wouter Verbeke <Wouter.Verbeke@vub.be
- Analytics for business applications
- Fraud detection
- Supply chain analytics
- Credit risk modeling
- Marketing analytics (customer churn prediction, netlift modeling, response modeling)
- Predictive and descriptive analytics
- Social and complex network analysis
- Profit driven data mining
Prof. Tias Guns <Tias.Guns@vub.be
- Exploratory data analysis
- Constraint-based mining
- Interactive pattern discovery
- Data summarization
- Prescriptive analytics: combining machine learning and optimisation
- Constraint Programming and discrete optimisation
- Demand sensing: long term industrial research collaboration to develop dynamic forecasting approaches for short term estimation of demand evolutions. The developed methods, based on ensemble approaches, are developed in the context of their integration into production planning systems.
- Uplift modeling: research collaboration with a financial institution to develop and investigate the usefulness of net-, true- or up- lift modeling approaches, to maximize returns of marketing campaigns by selecting the most appropriate channels, level of marketing intensity, etc.
- Fraud Analytics: research project aiming to make fraud detection models costsensitive. Development in collaboration with an industrial credit card company as part of a larger project aiming at combining artificial intelligence, cryptography, and data analytics for developing an intelligent, adaptive authorization system.