DocumentCode :
1866966
Title :
FC-MST: Feature correlation maximum spanning tree for multimedia concept classification
Author :
Hsin-Yu Ha ; Shu-Ching Chen ; Min Chen
Author_Institution :
Sch. of Comput. & Inf. Sci., Florida Int. Univ., Miami, FL, USA
fYear :
2015
fDate :
7-9 Feb. 2015
Firstpage :
276
Lastpage :
283
Abstract :
Feature selection is an actively researched topic in varies domains, mainly owing to its ability in greatly reducing feature space and associated computational time. Given the explosive growth of high-dimensional multimedia data, a well-designed feature selection method can be leveraged in classifying multimedia contents into high-level semantic concepts. In this paper we present a multi-phase feature selection method using maximum spanning tree built from feature correlation among multiple modalities (FC-MST). The method aims to first thoroughly explore not only the correlation between features within and across modalities, but also the association of features towards semantic concepts. Secondly, with the correlations, we identify important features and exclude redundant or irrelevant ones. The proposed method is tested on a well-known benchmark multimedia data set called NUS-WIDE and the experimental results show that it outperforms four well-known feature selection methods in all three important measurement metrics.
Keywords :
feature selection; multimedia systems; pattern classification; trees (mathematics); FC-MST; NUS-WIDE; feature correlation; feature space; high-dimensional multimedia data; high-level semantic concepts; maximum spanning tree; measurement metrics; multimedia concept classification; multimedia contents classification; multimedia data set; multiphase feature selection; multiple modalities; Correlation; Irrigation; Multimedia communication; Reliability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Semantic Computing (ICSC), 2015 IEEE International Conference on
Conference_Location :
Anaheim, CA
Type :
conf
DOI :
10.1109/ICOSC.2015.7050820
Filename :
7050820
Link To Document :
بازگشت