Coupled Clustering: a Method for Detecting Structural Correspondence

TL;DR

This paper introduces coupled clustering, a novel method for identifying structural correspondences between sub-structures of different systems, demonstrated on textual data to find topical links.

Contribution

It presents a new coupled clustering framework that simultaneously finds corresponding clusters across two datasets, advancing structural correspondence detection methods.

Findings

Effective in detecting topical correspondences in textual corpora

Demonstrates the viability of coupled clustering for structural analysis

Provides a new computational approach for cross-system cluster matching

Abstract

This paper proposes a new paradigm and computational framework for identification of correspondences between sub-structures of distinct composite systems. For this, we define and investigate a variant of traditional data clustering, termed coupled clustering, which simultaneously identifies corresponding clusters within two data sets. The presented method is demonstrated and evaluated for detecting topical correspondences in textual corpora.