An unsupervised content based news personalization using geolocation information

This paper presents content based recommendation system for news articles by filtering redundant news articles which describes the same event and happening from different sources by implementing unsupervised learning technique. A representative of similar articles is retained for further learning process. In this paper, a method for ranking news articles is also proposed based on similarity between user preference and the articles. Further the ranking is enhanced or improved by invoking location details of the user as they are more likely to be interested in their local news and event.