DocumentCode :
1678691
Title :
Active Learning for Semi-Supervised K-Means Clustering
Author :
Vu, Viet-Vu ; Labroche, Nicolas ; Bouchon-Meunier, Bernadette
Author_Institution :
LIP6, Univ. Pierre et Marie Curie-Paris 6, Paris, France
Volume :
1
fYear :
2010
Firstpage :
12
Lastpage :
15
Abstract :
K-Means algorithm is one of the most used clustering algorithm for Knowledge Discovery in Data Mining. Seed based K-Means is the integration of a small set of labeled data (called seeds) to the K-Means algorithm to improve its performances and overcome its sensitivity to initial centers. These centers are, most of the time, generated at random or they are assumed to be available for each cluster. This paper introduces a new efficient algorithm for active seeds selection which relies on a Min-Max approach that favors the coverage of the whole dataset. Experiments conducted on artificial and real datasets show that, using our active seeds selection algorithm, each cluster contains at least one seed after a very small number of queries and thus helps reducing the number of iterations until convergence which is crucial in many KDD applications.
Keywords :
data mining; learning (artificial intelligence); pattern clustering; active learning; data mining; knowledge discovery algorithm; min-max approach; seed based k-means clustering; semi-supervised k-means clustering; Clustering algorithms; Complexity theory; Convergence; Data mining; Equations; Nearest neighbor searches; Semi-supervied clustering; active learning; seed;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Tools with Artificial Intelligence (ICTAI), 2010 22nd IEEE International Conference on
Conference_Location :
Arras
ISSN :
1082-3409
Print_ISBN :
978-1-4244-8817-9
Type :
conf
DOI :
10.1109/ICTAI.2010.11
Filename :
5670014
Link To Document :
بازگشت