Class: A nonparametric clustering algorithm
1976; Elsevier BV; Volume: 8; Issue: 3 Linguagem: Inglês
10.1016/0031-3203(76)90011-x
ISSN1873-5142
AutoresFrederick R. Fromm, Richard A. Northouse,
Tópico(s)Advanced Clustering Algorithms Research
ResumoThe paper describes a nonparametric method for clustering of large data problems. The algorithm based on the ISODATA technique, calculates all required thresholds from the actual data, thus eliminating a priori estimates. Empirical derivation of the set of rules for calculating these parameters is presented. Results of using the technique on a number of artificial and real data samples are discussed.
Referência(s)