Application au domaine bancaire

In this paper, an original approach to database summarization is applied to a massive data set provided by a bank marketing department. The summarization process is based on an incremental and hierarchical conceptual clustering algorithm, building a summary hierarchy from database records. Levels of the hierarchy provide some views with different granulari- ties over the entire database. Each summary describes part of the data set. Furthermore, the fuzzy set-based representation of summaries allows the system to ensure a strong robustness and accuracy regarding the well-known threshold effect of the crisp clustering methods. The sum- marization process is also supported by some background knowledge, providing a user-friendly vocabulary to describe summaries with a high-level semantics. Even though our method is not immediately concerned with computational performance, its low time and memory requirements makes it appropriate for large real-life databases. The scalability of the process is demonstrated through the application on a banking data set.