One-class Classification based Finance News Story Recommendation

In this paper, we proposed an importance evaluation method for finance news story recommendation based on one-class classification. Based on the “important news stories” which are generated automatically as our training corpus, we used the one-class classification approach to evaluate the importance of each finance news story. This research not only quantifies the importance of each finance news story successfully, also makes it possible for ranking the results in finance specialty search engine in an innovative way. We investigated on the influence of features number and threshold to the performance of three one-class classification method Roccihio, k-means and one-class SVM. As experimental results shows, the method k-mean algorithm which had the best performance produced the precision up to 80% while maintaining recall at 95%.