A new active learning strategy in nearest neighbor classifier

摘要

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.