Vouw: Geometric Pattern Mining Using the MDL Principle

Lecture notes in computer science, pp. 158–170

Abstract

We introduce geometric pattern mining, the problem of finding recurring local structure in discrete, geometric matrices. It differs from existing pattern mining problems by identifying complex spatial relations between elements, resulting in arbitrarily shaped patterns. After we formalise this new type of pattern mining, we propose an approach to selecting a set of patterns using the Minimum Description Length principle. We demonstrate the potential of our approach by introducing Vouw, a heuristic algorithm for mining exact geometric patterns. We show that Vouw delivers high-quality results with a synthetic benchmark.

Authors 2

1. Micky Faas corresponding Aachen [LIACS

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, The Netherlands

   [LIACS, Leiden University, Leiden, The Netherlands]

2. Matthijs van Leeuwen Aachen [LIACS

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, The Netherlands

   [LIACS, Leiden University, Leiden, The Netherlands]