DocumentCode :
2918518
Title :
An Unsupervised Feature Selection Algorithm: Laplacian Score Combined with Distance-Based Entropy Measure
Author :
Liu, Rongye ; Yang, Ning ; Ding, Xiangqian ; Ma, Lintao
Author_Institution :
Dept. of Inf. Sci. & Eng., Ocean Univ. of China, Qingdao, China
Volume :
3
fYear :
2009
fDate :
21-22 Nov. 2009
Firstpage :
65
Lastpage :
68
Abstract :
In unsupervised learning paradigm, we are not given class labels, which features should we keep? Unsupervised feature selection method well solves this problem and has got a good effect in features selection with unlabeled data. Laplacian Score (LS) is a newly proposed unsupervised feature selection algorithm. However it uses k-means clustering method to select the top k features, therefore, the disadvantages of k-means clustering method greatly affect the result and increases the complexity of LS. In this paper, we introduce a novel algorithm called LSE (Laplacian Score combined with distance-based entropy measure) for automatically selecting subset of features. LSE uses distance-based entropy to replace the k-means clustering method in LS, which intrinsically solves the drawbacks of LS and contribute to the stability and efficiency of LSE. We compare LSE with LS on six UCI data sets. Experimental results demonstrate LSE can outperform LS on stability and efficiency, especially when processing high dimension datasets.
Keywords :
unsupervised learning; Laplacian score; distance-based entropy measure; k-means clustering method; unsupervised feature selection algorithm; unsupervised learning paradigm; Clustering algorithms; Clustering methods; Entropy; Filters; Information science; Information technology; Laplace equations; Oceans; Sea measurements; Stability; Distance-based Entropy Measure; LSE; Laplacian Score(LS); Unsupervised Feature Selction;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Information Technology Application, 2009. IITA 2009. Third International Symposium on
Conference_Location :
Nanchang
Print_ISBN :
978-0-7695-3859-4
Type :
conf
DOI :
10.1109/IITA.2009.390
Filename :
5369495
Link To Document :
بازگشت