Work

Numerical Pattern Mining Through Compression

Year: 2019

Type: conference-paper

Abstract: Pattern Mining (PM) has a prominent place in Data Science and finds its application in a wide range of domains. To avoid the exponential explosion of patterns different methods have been proposed. They are based on assumptions on interestingness and usually return very different pattern sets. In this paper we propose to use a compression-based objective as a well-justified and robust interestingness measure. We define the description lengths for datasets and use the Minimum Description Length principle (MDL) to find patterns that ensure the best compression. Our experiments show that the application of MDL to numerical data provides a small and characteristic subsets of patterns describing data in a compact way.

Source: DCC

DOI: 10.1109/dcc.2019.00019

Authors: Tatiana Makhalova, Sergei O. Kuznetsov, Amedeo Napoli

Institutions: National Research University Higher School of Economics, Laboratoire Lorrain de Recherche en Informatique et ses Applications, Centre Inria de l'Université de Lorraine, Université de Lorraine

Language: English

Topic: Data Mining Algorithms and Applications

Subfield: Information Systems

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W2946389537

FWCI: 0.841

Cites: 18

Cited by: 6

Related to: 10