Research on Cold-Start Problem in User Based Collaborative Filtering Algorithm

  • Lu LiuEmail author
  • Zhiqian Wang
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 905)


In order to solve the cold-start problem existing in traditional user based collaborative filtering algorithm, we propose a novel user clustering based algorithm, which firstly prefills user-item rating matrix, and then considers user characteristics as well as ratings when computing user similarities, and applies optimized k-means algorithm to cluster users. MovieLens is used as the test dataset. It is proved that the algorithm proposed in this paper can solve the cold-start problem and improve the accuracy of recommendation to some extent.


Cold-start Collaborative filtering K-means user clustering Similarity calculation optimization 


  1. 1.
    Wang, Q., Xie, Y., Yu, S.: Collaborative filtering algorithm combined with the user clustering and item types. Appl. Comput. Syst. 25(12), 132–137 (2016)Google Scholar
  2. 2.
    Liao, F., Lin, S., Guo, K.: A cold-starting personalized recommendation algorithm. J. Chin. Comput. Syst. 36(8), 1723–1727 (2015). (in Chinese)Google Scholar
  3. 3.
    Liu, H., Jing, L., Yu, J.: Survey of matrix factorization based social recommendation methods. J. Softw. 29(2), 1–24 (2018). (in Chinese)zbMATHGoogle Scholar
  4. 4.
    Xu, J., Yao, Y., Tong, H., Tao, X., Lu, J.: Ice-breaking: mitigating cold-start recommendation problem by rating comparison. In: International Conference on Artificial Intelligence, pp. 3981–3987 (2015)Google Scholar
  5. 5.
    Cao, C., Ni, Q., Zhai, Y.: An improved collaborative filtering recommendation algorithm based on community detection in social networks. In: Conference on Genetic & Evolutionary Computation, pp. 1–8 (2015)Google Scholar
  6. 6.
    Hu, X., Lu, H., Chen, X., Zhou, G.: K-means item clustering recommendation algorithm based on initial clustering center optimization. J. Air Force Early Warn. Acad. 28(3), 203–207 (2014). (in Chinese)Google Scholar
  7. 7.
    Ren, C.: Improved algorithm of alleviating item cold starting. Comput. Eng. Softw. 37(8), 11–15 (2016). (in Chinese)Google Scholar
  8. 8.
    Yu, H., Li, J.: Algorithm to solve the cold-start problem in new item recommendations. J. Softw. 26(6), 1395–1408 (2015). (in Chinese)MathSciNetGoogle Scholar
  9. 9.
    Ji, Z., Ge, W.: Research and optimization of k-means clustering algorithm in network recommendation systems. Inf. Commun. 175, 34–36 (2017). (in Chinese)Google Scholar
  10. 10.
    Bouras, C., Tsogkas, V.: Clustering to deal with the new user problem. In: IEEE 15th International Conference in Computational Science and Engineering (2012)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Beijing University of Posts and TelecommunicationsBeijingChina

Personalised recommendations