Approaches for measuring the stability of clustering methods
- 1 July 1982
- journal article
- Published by Association for Computing Machinery (ACM) in ACM SIGIR Forum
- Vol. 17 (1) , 6-20
- https://doi.org/10.1145/1095460.1095461
Abstract
Among the significant factors in assessing the suitability of a clustering technique to a given application is its stability; that is, how sensitive the algorithm is to perturbations in the input data. A number of techniques that appear to be suitable for measuring the stability of clustering have been published in the literature. When these techniques are closely examined, a number of generic approaches emerge. This note reviews these approaches and provides a classification of the various techniques appearing in the literature in terms of the approaches identified.Keywords
This publication has 19 references indexed in Scilit:
- Validity of clusters formed by graph-theoretic cluster methodsMathematical Biosciences, 1977
- The stability of two common matching functions in classification with respect to a proposed measureJournal of the American Society for Information Science, 1976
- Methods of Comparing ClassificationsAnnual Review of Ecology and Systematics, 1974
- A clustering algorithm based on user queriesJournal of the American Society for Information Science, 1974
- Metrics on spaces of finite treesJournal of Mathematical Psychology, 1973
- Objective Criteria for the Evaluation of Clustering MethodsJournal of the American Statistical Association, 1971
- A Review of ClassificationJournal of the Royal Statistical Society. Series A (General), 1971
- A Successive Approximations Approach to Character WeightingSystematic Zoology, 1969
- THE COMPARISON OF DENDROGRAMS BY OBJECTIVE METHODSTaxon, 1962
- A NEW MEASURE OF RANK CORRELATIONBiometrika, 1938