A Hybrid Framework for Improving Diversity and Long Tail Items in Recommendations
In today’s information overloaded era, recommender system is a necessity and it is widely used in most of the domains of e-commerce. Over the years, recommender system is improved to meet the main purpose of achieving better user experience, where accuracy is considered as one of the important aspects in its design. However, other aspects such as diversity, long tail item recommendation, novelty and serendipity are equally important while providing recommendations to the users. Research to improve above mentioned aspects is limited. In this paper, we propose an efficient approach to improve diversity and long tail item recommendations. The experiments are conducted on two real world movie rating datasets namely, MovieLens and Netflix. Experimental analysis shows that the proposed method outperforms the state-of-the art approaches in recommending diverse and long tail items.
KeywordsRecommender system Diversity Collaborative Filtering Long tail items Hybrid Reranking Framework
This work is partially funded by SERB, Department of Science and Technology, Govt. of India, Grant No. EMR/2017/004357, dated: 18th June 2018.
- 5.Harald, S.: Item popularity and recommendation accuracy. In: Proceedings of the Fifth ACM Conference on Recommender Systems, pp. 125–132 (2011)Google Scholar
- 10.Paterek, A.: Improving regularized singular value decomposition for collaborative filtering. In: Proceeding of KDD Cup Workshop at 13th ACM International Conference on Knowledge Discovery and Data Mining, pp. 39–42 (2007)Google Scholar
- 11.Pathak, A., Patra, B.K.: A knowledge reuse framework for improving novelty and diversity in recommendations. In: Proceedings of the Second ACM IKDD Conference on Data Sciences, pp. 11–19 (2015)Google Scholar
- 12.Sarwar, B., Karypis, G., Konstan, J., Riedl, J.: Item-based collaborative filtering recommendation algorithms. In: Proceedings of the 10th International Conference on World Wide Web, pp. 285–295 (2001)Google Scholar
- 13.Shanfeng, W., Maoguo, G., Haoliang, L., Junwei, Y.: A knowledge reuse framework for improving novelty and diversity in recommendations. In: Proceedings of the Second ACM IKDD Conference on Data Sciences, vol. 104, pp. 145–155 (2016)Google Scholar