Propositionalisation and Aggregates

A.J. Knobbe, M. de Haas, and A. Siebes. In Proceedings of the 5th European Conference on Principles of Data Mining and Knowledge Discovery, pages 277--288. Springer-Verlag, September 2001. More behind this link.

Abstract

The fact that data is scattered over many tables causes many problems in the practice of data mining. To deal with this problem, one either constructs a single table by hand, or one uses a Multi-Relational Data Mining algorithm. In this paper, we propose a different approach in which the single table is constructed automatically using aggregate functions, which repeatedly summarise information from different tables over associations in the datamodel. Following the construction of the single table, we apply traditional data mining algorithms. Next to an in-depth discussion of our approach, the paper presents results of experiments on three well-known data sets.

BibTeX entry.

Propositionalisation and Aggregates

Abstract

Other publications