Open AccessInformation-Theoretic Based Clustering Method for High-Dimensional Data
Author(s) -
Xuan Huang,
Chen Lixi,
Yinsong Ye
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1533/2/022115
Subject(s) - cluster analysis , cure data clustering algorithm , correlation clustering , clustering high dimensional data , data mining , computer science , data stream clustering , consensus clustering , fuzzy clustering , single linkage clustering , canopy clustering algorithm , pattern recognition (psychology) , artificial intelligence
Clustering is one of the important techniques for pattern recognition. Traditional clustering methods often group data objects by distance measurement. However, it is difficult to measure data in high-dimensional space. This paper proposes a clustering method for high-dimensional data. It combines the information theory criteria to establish clustering rules. The improved of K-Means is used to generate basis clustering, then the clustering ensemble is used to integrate the initial clusters to obtain a more stable final result. Our proposed method is tested on standard datasets, and its performance is compared with K-means ensemble. The experimental results indicate effectiveness for high-dimensional data clustering.