Keyword-Labeled Classification with Auxiliary Unlabeled Documents

To reduce the human effort in labeling the training set for document classification, some learning algorithms ask users to give the representative keywords for each class rather than any labeled documents. The key challenge in such emph {keyword-labeled classification} is how to learn the high quality classifier with very small number of keywords. In this paper, we propose a novel co-clustering based classification algorithm for keyword-labeled classification (CCKC) by utilizing auxiliary unlabeled documents. The experimental results show our algorithm greatly improves the classification performance over existing approaches.