DocumentCode :
2771153
Title :
Combining Diversity and Classification Accuracy for Ensemble Selection in Random Subspaces
Author :
Ko, Albert Hung-Ren ; Sabourin, Robert ; de Souza Britt, A.
Author_Institution :
Quebec Univ., Montreal
fYear :
0
fDate :
0-0 0
Firstpage :
2144
Lastpage :
2151
Abstract :
An ensemble of classifiers has been shown to be effective in improving classifier performance. Two elements are believed to be viable in constructing an ensemble: a) the classification accuracy of each individual classifier; and b) diversity among the classifiers. Nevertheless, most works based on diversity suggest that there exists only weak correlation between diversity and ensemble accuracy. We propose compound diversity functions which combine the diversities with the classification accuracy of each individual classifier, and show that with Random subspaces ensemble creation method, there is a strong correlation between the proposed functions and ensemble accuracy. The statistical result indicates that compound diversity functions perform better than traditional diversity measures.
Keywords :
pattern classification; statistical analysis; classification accuracy; ensemble creation method; ensemble selection; random subspaces; Bagging; Boosting; Diversity reception; Error correction; Pattern recognition; Performance evaluation; System testing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 2006. IJCNN '06. International Joint Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-9490-9
Type :
conf
DOI :
10.1109/IJCNN.2006.246986
Filename :
1716376
Link To Document :
بازگشت