Abstract
Abstract. A novel fast and accurate supervised learning algorithm is proposed as a general text classification algorithm for linearly separated data. The strategy of the algorithm takes advantage of the training errors to successively refine an initial classifier. Experimental evaluation of the proposed algorithm on standard text collections, show that results compared favorably to those from state of the art algorithms such as SVMs. Experiments conducted on the datasets provided in the framework of the ECDL/PKDD 2008 Challenge for Spam Detection in Social Bookmarking Systems, demonstrate the effectiveness of the proposed algorithm. 1
Users
Please
log in to take part in the discussion (add own reviews or comments).