A Split-Merge Framework for Comparing Clusterings
Abstract
Clustering evaluation measures are frequently used to evaluate the performance of algorithms. However, most measures are not properly normalized and ignore some information in the inherent structure of clusterings. We model the relation between two clusterings as a bipartite graph and propose a general component-based decomposition formula based on the components of the graph. Most existing measures are examples of this formula. In order to satisfy consistency in the component, we further propose a split-merge framework for comparing clusterings of different data sets. Our framework gives measures that are conditionally normalized, and it can make use of data point information, such as feature vectors and pairwise distances. We use an entropy-based instance of the framework and a coreference resolution data set to demonstrate empirically the utility of our framework over other measures.
Cite
@article{arxiv.1206.6475,
title = {A Split-Merge Framework for Comparing Clusterings},
author = {Qiaoliang Xiang and Qi Mao and Kian Ming Chai and Hai Leong Chieu and Ivor Tsang and Zhendong Zhao},
journal= {arXiv preprint arXiv:1206.6475},
year = {2012}
}
Comments
Appears in Proceedings of the 29th International Conference on Machine Learning (ICML 2012)