Advertisement

Improving Jaccard Index Using Genetic Algorithms for Collaborative Filtering

  • Soojung LeeEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10385)

Abstract

As data sparsity may produce unreliable recommendations in collaborative filtering-based recommender systems, it has been addressed by many researchers in related fields. Jaccard index is regarded as effective when combined with existing similarity measures to relieve data sparsity problem. However, the index only reflects how many items are co-rated by two users, without considering whether their ratings are evaluated similar or not. This paper proposes a novel improvement of Jaccard index, reflecting not only the ratio of co-rated items but also whether the ratings of each co-rated item by two users are both high, medium, or low. A genetic algorithm is employed to find the optimal weights of the levels of evaluations and the optimal boundaries between them. We conducted extensive experiments to find that the proposed index significantly outperforms Jaccard index on moderately sparse to dense datasets, in terms of both prediction and recommendation qualities.

Keywords

Similarity measure Jaccard coefficient Collaborative filtering Recommender system 

References

  1. 1.
    Aamir, M., Bhusry, M.: Recommendation system: state of the art approach. Int. J. Comput. Appl. 120(12), 25–32 (2015)Google Scholar
  2. 2.
    Ahn, H.J.: A new similarity measure for collaborative filtering to alleviate the new user cold-starting problem. Inf. Sci. 178(1), 37–51 (2008)CrossRefGoogle Scholar
  3. 3.
    Bobadilla, J., Serradilla, F., Bernal, J.: A new collaborative filtering metric that improves the behavior of recommender systems. Knowl.-Based Syst. 23(6), 520–528 (2010)CrossRefGoogle Scholar
  4. 4.
    Bobadilla, J., Ortega, F., Hernando, A., Bernal, J.: A collaborative filtering approach to mitigate the new user cold start problem. Knowl.-Based Syst. 26, 225–238 (2012)CrossRefGoogle Scholar
  5. 5.
    Bobadilla, J., Ortega, F., Hernando, A., Gutierrez, A.: Recommender systems survey. Knowl.-Based Syst. 46, 109–132 (2013)CrossRefGoogle Scholar
  6. 6.
    Jamali, M., Ester, M.: Trustwalker: a random walk model for combining trust-based and item-based recommendation. In: 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 397–406. ACM (2009)Google Scholar
  7. 7.
    Koutrica, G., Bercovitz, B., Garcia-Molina, H.: FlexRecs: expressing and combining flexible recommendations. In: The 2009 ACM SIGMOD International Conference Management of Data, pp. 745–758. ACM (2009)Google Scholar
  8. 8.
    Liu, H., Hu, Z., Mian, A., Tian, H., Zhu, X.: A new user similarity model to improve the accuracy of collaborative filtering. Knowl.-Based Syst. 56, 156–166 (2014)CrossRefGoogle Scholar
  9. 9.
    Ren, L., Gu, J., Xia, W.: A weighted similarity-boosted collaborative filtering approach. Energy Procedia 13, 9060–9067 (2011)CrossRefGoogle Scholar
  10. 10.
    Resnick, P., Lakovou, N., Sushak, M., Bergstrom, P., Riedl, J.: Grouplens: an open architecture for collaborative filtering of netnews. In: Proceedings the ACM Conference on Computer Supported Cooperative Work, pp. 175–186. ACM Press (1994)Google Scholar
  11. 11.
    Saranya, K.G., Sadasivam, G.S., Chandralekha, M.: Performance comparison of different similarity measures for collaborative filtering technique. Indian J. Sci. Technol. 9(29) (2016)Google Scholar
  12. 12.
    Su, X., Khoshgoftaar, T.M.: A survey of collaborative filtering techniques. Adv. Artif. Intell. 2009, 4 (2009)CrossRefGoogle Scholar
  13. 13.
    Sun, H.-F., et al.: JacUOD: a new similarity measurement for collaborative filtering. J. Comput. Sci. Technol. 27(6), 1252–1260 (2012)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Gyeongin National University of EducationAnyangKorea

Personalised recommendations