Closed pattern mining of interval data and distributional data

TL;DR

This paper introduces pattern languages for mining closed patterns in interval and distributional data, enabling applications in clustering and supervised learning through encoding as itemsets.

Contribution

It presents novel pattern languages based on intersection and inclusion constraints for interval data, extending to distributional data, and demonstrates their use in machine learning tasks.

Findings

Effective encoding of interval patterns as itemsets

Successful application in clustering tasks

Extension to distributional data

Abstract

We discuss pattern languages for closed pattern mining and learning of interval data and distributional data. We first introduce pattern languages relying on pairs of intersection-based constraints or pairs of inclusion based constraints, or both, applied to intervals. We discuss the encoding of such interval patterns as itemsets thus allowing to use closed itemsets mining and formal concept analysis programs. We experiment these languages on clustering and supervised learning tasks. Then we show how to extend the approach to address distributional data.