Recommender systems are turning into increasingly important to individual users and businesses for providing customized recommendations. However, while the bulk of algorithms proposed in recommender systems literature have targeted on improving recommendation accuracy (as exemplified by the recent Netflix Prize competition), other necessary aspects of recommendation quality, like the diversity of recommendations, have usually been overlooked. In this paper, we have a tendency to introduce and explore a number of item ranking techniques that may generate recommendations that have substantially higher combination diversity across all users while maintaining comparable levels of advice accuracy. Comprehensive empirical evaluation consistently shows the variety gains of the proposed techniques using several real-world rating datasets and completely different rating prediction algorithms.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here