Recommender systems are becoming increasingly important to individual users and businesses for providing personalized recommendations. However, while the majority of algorithms proposed in recommender systems literature have focused on improving recommendation accuracy, other important aspects of recommendation quality, such as the diversity of recommendations, have often been overlooked. In this paper, we introduce and explore a number of item ranking techniques that can generate recommendations that have substantially higher aggregate diversity across all users while maintaining comparable levels of recommendation accuracy. Comprehensive empirical evaluation consistently shows the diversity gains of the proposed techniques using several real-world rating datasets and different rating prediction algorithms. Improving Aggregate Recommendation Diversity Using Ranking-Based Techniques
There is a growing awareness of the importance of aggregate diversity in recommender systems. Furthermore, while, as mentioned earlier, there has been significant amount of work done on improving individual diversity, the issue of aggregate diversity in recommender systems has been largely untouched.
In real world settings, recommender systems generally perform the following two tasks in order to provide recommendations to each user. First, the ratings of unrated items are estimated based on the available information (typically using known user ratings and possibly also information about item content or user demographics) using some recommendation algorithm. And second, the system finds items that maximize the user’s utility based on the predicted ratings, and recommends them to the user. Ranking approaches proposed in this paper are designed to improve the recommendation diversity in the second task of finding the best items for each user.