An Evaluation of Criteria for Measuring the Quality of Clusters
Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence. Natural Language Processing. Pages 905-909.
Abstract
An important problem in clustering is how to decide what is the best set of clusters for a given data set, in terms of both the number of clusters and the membership of those clusters. In this paper we develop four criteria for measuring the quality of different sets of clusters. These criteria are designed so that different criteria prefer cluster sets that generalise at different levels of granularity. We evaluate the suitability of these criteria for non-hierarchical clustering of the results returned by a search engine. We also compare the number of clusters chosen by these criteria with the number of clusters chosen by a group of human subjects. Our results demonstrate that our criteria match the variability exhibited by human subjects, indicating there is no single perfect criterion. Instead, it is necessary to select the correct criterion to match a human subject's generalisation needs.
@inproceedings{ijcai1999p129,
title = {An Evaluation of Criteria for Measuring the Quality of Clusters},
author = {Raskutti, Bhavani and Leckie, Christopher},
booktitle = {Proceedings of the Sixteenth International Joint Conference on
Artificial Intelligence, {IJCAI-99}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Thomas Dean},
pages = {905--909},
year = {1999},
note = {Natural Language Processing},
url = {https://ijcai.org/proceedings/1999/129},
}
TY - CPAPER TI - An Evaluation of Criteria for Measuring the Quality of Clusters AU - Raskutti, Bhavani AU - Leckie, Christopher T2 - Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence, IJCAI-99 PB - International Joint Conferences on Artificial Intelligence PY - 1999 SP - 905 EP - 909 UR - https://ijcai.org/proceedings/1999/129 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
