Abstract. In this paper, we discuss the development of a hybrid multistrategy book recommendation system using Linked Open Data. Our approach builds on training individual base recommenders and using global popularity scores as generic recommenders. The results of the individual recommenders are combined using stacking regression and rank aggregation. We show that this approach delivers very good results in different recommendation settings and also allows for incorporating diversity of recommendations.Keywords: Linked Open Data, Hybrid Recommender Systems, Stacking
Overall ApproachWe propose a hybrid, multi-strategy approach that combines the results of different base recommenders and generic recommenders into a final recommendation. A base recommender is an individual collaborative or content based recommender system, whereas a generic recommender makes a recommendation solely on some global popularity score, which is the same for all users. For base recommenders, we use two collaborative filtering strategies (item and user based), as well as different content-based strategies exploiting various feature sets created from DBpedia 1 . Our results in the three tasks of the LOD-enabled Recommender Systems Challenge 2014 2 demonstrate the effectiveness of our approach.
Generic RecommendersWe use different generic recommenders in our approach. First, the RDF Book Mashup dataset 3 provides the average score assigned to a book on Amazon. Furthermore, DBpedia provides the number of ingoing links to the Wikipedia 1 http://dbpedia.org 2 75,559 numeric ratings on 6,166 books (from 0-5, Task 1) and 72,372 binary ratings on 6733 books (Tasks 2 and 3), resp., from 6,181 users for training, and evaluation on 65,560 and 67,990 unknown ratings, resp. See http://challenges.2014.eswcconferences.org/index.php/RecSys for details. 3