Title :
A new active learning strategy in nearest neighbor classifier
Author :
Wei-Ran Song ; Cai, Yong-hua ; Wu, Bo ; Sun, Tao
Author_Institution :
Coll. of Appl. Sci., Beijing Univ. of Technol., Beijing, China
Abstract :
In this paper, we propose an active sample selection algorithm (SSME) based on maximum entropy criterion. By calculating the information entropy of the unlabeled samples, the algorithm can find the most informative samples from unlabeled data set. Comparative experiments with random selection algorithm are conducted on 10 real data sets. The results show the superiority of our proposed algorithm in terms of predictive accuracy and condensing rate.
Keywords :
entropy; learning (artificial intelligence); pattern classification; SSME; active learning strategy; active sample selection algorithm; condensing rate; information entropy; maximum entropy criterion; nearest neighbor classifier; predictive accuracy; Abstracts; Glass; Heart; Iris; Active learning; Instance selection; Maximum entropy criterion; Nearest neighbor classifier;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2012 International Conference on
Conference_Location :
Xian
Print_ISBN :
978-1-4673-1484-8
DOI :
10.1109/ICMLC.2012.6359015
