Numerical Pattern Mining Through Compression - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

Numerical Pattern Mining Through Compression

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.
Fichier principal
Vignette du fichier
DCC_makhalova.pdf (605.29 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02162927 , version 1 (23-06-2019)
hal-02162927 , version 2 (29-10-2020)

Identifiers

Cite

Tatiana Makhalova, Sergei O. Kuznetsov, Amedeo Napoli. Numerical Pattern Mining Through Compression. DCC 2019 - 2019 Data Compression Conference, Mar 2019, Snowbird, United States. pp.112-121, ⟨10.1109/DCC.2019.00019⟩. ⟨hal-02162927v2⟩
109 View
221 Download

Altmetric

Share

Gmail Facebook X LinkedIn More