Collaborative filtering is present within current web-based systems in many forms. At the beginning they were mostly either item based or user based, but as the time passed, many hybrid approaches combining several techniques from multiple disciplines emerged. However, the basic idea remained always the same: use past experiences of users to get benefits for an individual. We can imagine it as when we are in the woods, we take the paths that others took before us. Significant enhancement of the basic idea is not to use past experiences of all users but instead of it, consider only users which are similar to the user, for who we are targeting recommendations.

Based on this we can start to detect virtual communities of users. As collaborative filtering is always tailored to the specific domain, we chose to focus on recommending news articles, a highly dynamic domain with frequent changes and special user behavior. Nobody reads yesterdays newspaper. Often by picking certain articles, we indirectly express our opinions and preferences. In our work, we consider these fluctuating and time-dependent changes by incorporating influence of volatile communities on recommendations. We use keyword-based layered user models, where layers represent different attributes (e.g. articles from a week ago, long term preference). For the semantics we use latent semantic model or WordNet, the actual keywords are extracted from the plain texts of articles. We assume that article categories are represented by different keywords, so a model can be partitioned based on clusters of keywords corresponding to those categories. The novel approach is to create virtual communities based on these clusters. A user can belong to several communities that are on the same hierarchical level. Recommendations coming from such communities would be more accurate than when we group users based on whole user profile.