Abstract | ||
---|---|---|
Evidential clustering algorithms produce credal partitions that enhance the concepts of hard, fuzzy or possibilistic partitions to represent all assignments ranging from complete ignorance to total certainty. This paper introduces the first semi-supervised extension of the evidential c-means clustering algorithm that can benefit from the introduction of a small set of labeled data (or seeds). Experiments conducted on real datasets show that the introduction of seeds can lead to a significant increase in clustering accuracy compared to a traditional evidential clustering algorithm as well as a decrease in the number of iterations to convergence. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/SCIS-ISIS.2014.7044676 | SCIS&ISIS |
Keywords | Field | DocType |
fuzzy set theory,pattern clustering,possibility theory,credal partitions,evidential c-means clustering algorithm,evidential seed-based semisupervised clustering,fuzzy partition,hard partition,labeled data,possibilistic partition | Canopy clustering algorithm,Fuzzy clustering,CURE data clustering algorithm,Data stream clustering,Pattern recognition,Correlation clustering,Computer science,Constrained clustering,Artificial intelligence,Cluster analysis,Machine learning,Single-linkage clustering | Conference |
ISSN | Citations | PageRank |
2377-6870 | 1 | 0.36 |
References | Authors | |
22 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Antoine, V. | 1 | 2 | 1.04 |
Labroche, N. | 2 | 2 | 1.72 |
Viet-Vu Vu | 3 | 1 | 0.36 |