Dynamic Bayesian modeling for risk prediction in credit operations

Hanen Borchani, Ana Maria Martinez, Andres Masegosa, Helge Langseth, Thomas Dyhre Nielsen, Antonio Salmerón, Antonio Fernández, Anders Læsø Madsen, Ramón Sáez

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

4 Citations (Scopus)
333 Downloads (Pure)


Our goal is to do risk prediction in credit operations, and as data is collected continuously and reported on a monthly basis, this gives rise to a streaming data classification problem. Our analysis reveals some practical problems that have not previously been thoroughly analyzed in the context of streaming data analysis: the class labels are not immediately available and the relevant predictive features and entities under study (in this case the set of customers) may vary over time. In order to address these problems, we propose to use a dynamic classifier with a wrapper feature subset selection to find relevant features at different time steps. The proposed model is a special case of a more general framework that can also accommodate more expressive models containing latent variables as well as more sophisticated feature selection schemes.
Original languageEnglish
Title of host publicationThe 13th Scandinavian Conference on Artificial Intelligence (SCAI'2015)
PublisherIOS Press
Publication date2015
ISBN (Print)978-1-61499-588-3
ISBN (Electronic)978-1-61499-589-0
Publication statusPublished - 2015
Event13th Scandinavian Conference on Artificial Intelligence - Halmstad University, Halmstad, Sweden
Duration: 4 Nov 20156 Nov 2015
Conference number: 13th


Conference13th Scandinavian Conference on Artificial Intelligence
LocationHalmstad University
SeriesFrontiers in Artificial Intelligence and Applications


  • Streaming data
  • Dynamic Bayesian networks
  • Variational Bayes
  • Feature subset selection
  • Credit operations


Dive into the research topics of 'Dynamic Bayesian modeling for risk prediction in credit operations'. Together they form a unique fingerprint.

Cite this