Investigations into User Rating Information and Predictive Accuracy in a Collaborative Filtering Domain

The work described in this paper extracts user feature values from three Collaborative Filtering datasets and uses a supervised machine learning approach to identify if there is any underlying relationship between these feature values and the average precision score obtained. The overall hypothesis is, if there is any underlying relationship between features and the precision score, by checking the values of these features for a user, we can know in advance of recommendation the level of accuracy we could expect from the collaborative filtering system. Thus, a user may have more (or less) confidence in the recommendations produced.