Enhancing News Articles Clustering using Word N-Grams

In this work we explore the possible enhancement of the document clustering results, and in particular clustering of news articles from the web, when using word-based n-grams during the keyword extraction phase. We present and evaluate a weighting approach that combines clustering of news articles derived from the web using n-grams, extracted from the articles at an offline stage. We compared this technique with the single minded bag-of-words representation that our clustering algorithm, W-kmeans, previously used. Our experimentation revealed that via tuning of the weighting parameters between keyword and n-grams, as well as the n itself, a significant improvement regarding the clustering results metrics can be achieved. This reflects more coherent clusters and better overall clustering performance.