Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

Probabilistic Reinterpretation of Collaborative Filtering Approaches Considering Cluster Information of Item Contents

항목 내용물의 클러스터 정보를 고려한 협력필터링 방법의 확률적 재해석

  • 김병만 (금오공과대학교 컴퓨터공학과) ;
  • 이경금 (오공과대학교 컴퓨터공학과) ;
  • 오상엽 (금오공과대학교 컴퓨터공학과)
  • Published : 2005.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

With the development of e-commerce and the proliferation of easily accessible information, information filtering has become a popular technique to prune large information spaces so that users are directed toward those items that best meet their needs and preferences. While many collaborative filtering systems have succeeded in capturing the similarities among users or items based on ratings to provide good recommendations, there are still some challenges for them to be more efficient, especially the user bias problem, non-transitive association problem and cold start problem. Those three problems impede us to capture more accurate similarities among users or items. In this paper, we provide probabilistic model approaches for UCHM and ICHM which are suggested to solve the addressed problems in hopes of achieving better performance. In this probabilistic model, objects (users or items) are classified into groups and predictions are made for users considering the Gaussian distribution of user ratings. Experiments on a real-word data set illustrate that our proposed approach is comparable with others.

인터넷의 상업적 이용이 증가하고 인터넷에서 쉽게 얻을 수 있는 정보의 양이 풍성해지면서 정보 필터링 (information filtering) 기법은 대량의 정보 공간에서 사용자의 요구와 기호에 맞는 항목을 찾는 과정에 널리 사용되고 있다. 많은 협력필터링 (collaborative filtering) 시스템이 사용자 평가를 기반으로 사용자나 항목들 사이의 유사성을 찾아내고 이를 바탕으로 추천을 해왔지만 사용자 편향 (user bias), 비전이 연관 (non-transitive association), cold start 문제와 같이 성능을 높이기 위해 해결해야 할 문제들이 남아있다. 이 세 가지 문제는 사용자나 항목들 사이에 더 정확한 유사도를 찾아내는 과정에 장애가 된다. 본 논문에서는 이러한 문제들을 해결하기 위해 제안된 UCHM 및 ICHM 방법을 확률적으로 재해석하였다. 이 확률적 모델은 객체 (사용자 또는 품목)들을 그룹들로 구분하고 각 그룹 내에서 사용자 평가가 가우시안 분포를 따른다는 가정 하에 사용자들이 무엇을 선호할 것인지 예측한다. 실세계 자료에 대한 실험 결과, 제안된 방식이 다른 방식들과 비교할 만한 성능을 보인다는 것을 확인할 수 있었다.

Keywords

References

  1. Resnick, P., Iacovou, N., Suchak, M., Bergstorm, P. and Riedl, J., 'GroupLens: An open architecture for collaborative filtering of Netnews,' Proc. of the ACM CSCW-94, pp. 175-186, 1994
  2. Upendra, S. and Patti, M., 'Social Information Filtering: Algorithms for Automating Word of Mouth,' Proc. of the ACM CHI'95 Conf. on Human Factors in Computing Systems, pp. 210-217, 1995 https://doi.org/10.1145/223904.223931
  3. Sarwar, B. M., Karypis, G., Konstan, J. A. and Riedl, J., 'Item-based Collaborative Filtering Recommendation Algorithms,' Proc. of the Tenth Int. WWW Conf. 2001, pp. 285-295, 2001 https://doi.org/10.1145/371920.372071
  4. Hofmann, T., 'Collaborative Filtering via Gaussian Probabilistic Latent Semantic Analysis,' Proc. of the SIGIR'03, pp. 259-266, 2003 https://doi.org/10.1145/860435.860483
  5. Jin, R., Luo Si, Chengxiang Zhai, James P. Callan, 'Collaborative filtering with decoupled models for preferences and rating,' Proc. of the CIKM 2003, pp.309-316, 2003 https://doi.org/10.1145/956863.956922
  6. O'Conner, M. and Herlocker, J., 'Clustering items for collaborative filtering,' Proc. of the ACM-SIGIR Workshop on Recommender Systems, 1999
  7. Popescul, A., Lyle H. Ungar, David M. Pennock, and Steve Lawrence, 'Probabilistic Models for Unified Collaborative and Content-Based Recommendation in Sparse-Data Environments,' Proc. of the 17'th Conference on UAI, 2001
  8. Si, Luo and Jin, Rong, 'Flexible Mixture Model for Collaborative Filtering,' Proc. of the ICML 2003, pp.704-711, 2003
  9. M. Balabanovic and Y. Shoham, 'Fab : Content-based collaborative recommendation,' CACM, Vol.40, No.3, 1997 https://doi.org/10.1145/245108.245124
  10. C. Basu, H. Hirsh, and W. Cohen, 'Recommendation as Classification : Using Social and Content-Based Information in recommendation,' Proc. of AAAI, 1998
  11. N. Good, J.B. Schafer, J.A. Konstan, A. Borchers, B. Sarwar, J. Herlocker and J. Riedl, 'Combininig Collaborative Filtering with Personal Agents for Better Recommendations,' Proc. of the AAAI-99, 1999
  12. Q. Li and B. M. Kim, 'Constructing User Profiles for Collaborative Recommender System,' Advanced Web Technologies and Applications: 6th Asia-Pacific Web Conference, J. X. Yu, X. Lin, H. Lu and Y. Zhang, eds., LNCS 3007, Springer-Verlag, pp. 100-110, 2004 https://doi.org/10.1007/b96838
  13. 김병만, 이경, 김시관, 임은기, 김주연 '추천시스템을 위한 내용기반 필터링과 협력필터링의 새로운 결합기법', 한국정보과학학회논문지: 소프트웨어및응용 31권 3호, pp.332-342, 2004
  14. Byeong Man Kim, Qing Li, Jong-Wan Kim and Jinsoo Kim, 'A New Collaborative Recommender System Addressing three Problems,' PRICAI 2004 : Trends in Artificial Intelligence, C. Zhang, H. W. Guesgen and W. K. Yeap, Eds., LNAI 3157, Springer-Verlag, pp.495-504, 2004 https://doi.org/10.1007/b99563
  15. 김병만, 이경, '항목 속성과 평가정보를 이용한 혼합 추천방법', 한국정보과학학회논문지 : 소프트웨어 및 응용 31권 12호, pp. 1672-1683, 2004
  16. Breese, J. S., Heckerman, D. and Kardie, C., 'Empirical Analysis of Predictive Algorithms for Collaborative Filtering,' Proc. Of the 14th UAI, pp. 43-52, 1998
  17. Fisher,D., Kris Hildrum, Jason Hong, Mark Newman, Megan Thomas, Rich Vuduc, 'SWAMI: a framework for collaborative filtering algorithm development and evaluation,' Proc. of the SIGIR 2000, pp.366-368, 2000 https://doi.org/10.1145/345508.345658
  18. Huang, Z., Hsinchun Chen and Daniel Zeng, 'Applying associative retrieval techniques to alleviate the sparsity problem in collaborative filtering,' ACM TOIS, Vol. 22, No.1, 2004 https://doi.org/10.1145/963770.963775
  19. Gupta, D., Digiovanni, M., H. Narita, K. Goldberg, 'Jester 2.0: Evaluation of an New Linear Time Collaborative Filtering Algorithm,' Proc. of the SIGIR-99, pp.291-292, 1999 https://doi.org/10.1145/312624.312718
  20. T. M. Mitchell, Machine Learning, McGRAW-HILL, 1997
  21. Han, J., and Kamber, M., Data mining: Concepts and Techniques. New York: Morgan-Kaufman, 2000
  22. Bradley, P. S. and Fayyad, U.M., 'Refining Initial Points for K-Means Clustering,' Proc. of the ICML '98, pp.91-99, 1998
  23. J. Herlocker, J. Konstan, A. Borchers, and J. Riedl, 'An Algorithmic Framework for Performing Collaborative Filtering,' In Proceedings of ACM SiGIR-99, 1999 https://doi.org/10.1145/312624.312682
  24. Nasraoui,O. and M. Pavuluri, 'Accurate Web Recommendations Based on Profile-Specific URL-Predictor Neural Networks,' Proc. of the WWW04, 2004
  25. Nasraoui,O. and M. Pavuluri, 'A Context Ultra-Sensitive Approach to High Quality Web Recommendations based on Web Usage Mining and Neural Network Committees,' Proc. of the International Web Dynamics Workshop of WWW04, 2004