Redescription Mining: An Overview.

Esther Galbrun (1, 2) , Pauli Miettinen (3)

1 ORPAILLEUR - Knowledge representation, reasonning
2 Aalto University
3 Max Planck Institute for Informatics [Saarbrücken]

In many real-world data analysis tasks, we have different types of data over the same objects or entities, perhaps because the data originate from distinct sources or are based on different terminologies. In order to understand such data, an intuitive approach is to identify the correspondences that exist between these different aspects. This is the motivating principle behind redescription mining, a data analysis task that aims at finding distinct common characterizations of the same objects. This paper provides a short overview of redescription mining; what it is and how it is connected to other data analysis methods; the basic principles behind current algorithms for redescription mining; and examples and applications of redescription mining for real-world data analysis problems.

Mots clés: redescription mining; alternative characterizations; visualizations; data mining; Index Terms—Redescription mining; alternative characteriza- tions; visualizations; data mining

Domaines: Intelligence artificielle [cs.AI]

Esther Galbrun, Pauli Miettinen. Redescription Mining: An Overview.. IEEE Intelligent Informatics Bulletin, 2017, 18 (2), pp.7-12. ⟨hal-01726074⟩