跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 巫孟倫
Meng-lun Wu
論文名稱： 基於共分群模型整合內容式與協同式之即時推薦系統
A scalable framework for integrating content-based filtering with collaborative filtering using co-clustering with augmented matrices
指導教授： 張嘉惠
Chia-hui Chang
口試委員:
學位類別： 博士
Doctor
系所名稱： 資訊電機學院 - 資訊工程學系
Department of Computer Science & Information Engineering
論文出版年： 2014
畢業學年度： 102
語文別： 英文
論文頁數： 94
中文關鍵詞： 協同式推薦系統 、內容式推薦系統 、共分群 、雲端運算
外文關鍵詞： Collaborative filtering, Content-based filtering, Co-clustering, Hadoop Map-Reduce
相關次數： 點閱：8  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 推薦系統是近年來相當熱門的研究主題。常見的做法包含協同式推薦系統及內容式推薦系統。協同式推薦系統受限於資料過於稀疏(data sparsity)以及冷啓動(cold start)兩個問題。內容式推薦系統則多侷限於使用者既有的興趣，較難針對推薦系統的目標函數做最佳化。內容式推薦系統雖然沒有協同式推薦系統的效果好，但是卻能夠幫助協同式推薦系統解決冷僻問題之窘境。因此也有許多研究，綜合兩者之方法，以截長補短的方式進行推薦。
    因此在這篇論文中，我們提出一混和模型，以CCAM共分群演算法整合協同式推薦系統及內容式推薦系統，以解決cold start以及data sparsity的問題。CCAM是一基於information theoretic co-clustering的共分群演算法，但考慮了額外的內容資訊，像是使用者的特性資料及產品的特徵等。最後，我們將此混和模型實作於Hadoop Map-Reduce平台，以期望發展一兼顧效率與效能之推薦系統。

    Recommender systems have become an essential research field because of a high interest from academia and industries. Collaborative filtering (CF), a branch of recommender system, is frequently confronted with the sparsity issue (resulted in fewer records (rating / clicking) against the unknowns that need to be predicted) and “cold start” problem (hard to make prediction for new user and new item), while Content-based (CB) approaches are limited by recommending similar items without user-item click information. Empirically, CF is better than CB, but is helpful to solve cold-start problem. Therefore, many hybrid approaches have been proposed to integrate collaborative filtering and content-based approach.
    In this thesis, we propose a hybrid approach that combines content-based approach with collaborative filtering under a unified model called co-clustering with augmented matrices (CCAM). CCAM is based on information theoretic co-clustering but further considers augmented matrices like user profile and item description. We then build a collaborative filtering model based on content-based information and co-clustering result to reduce the sparsity problem and solve cold-start problem. Finally, a parallel approach is proposed to solve the scalability problem of large data set.

    English Abstract i Chinese Abstract ii Contents iii List of Figures vi List of Tables viii 1 Introduction p.1 2 Related Works p.7 2.1 Co-clustering p.7 2.1.1 Evaluation of co-clustering algorithm p.9 2.2 Recommender system p.10 2.3 Parallel co-clustering approach for recommender system p.13 3 Co-clustering with Augmented Matrices (CCAM) p.16 3.1 Introduction p.16 3.2 Problem Definition p.17 3.3 Co-clustering with augmented matrices algorithm p.20 3.4 Illustration Example p.22 3.5 Evaluation of co-clustering result p.23 3.5.1 Data description p.23 3.5.2 Classification-oriented evaluation p.24 3.5.3 Mutual information based evaluation p.27 3.5.4 Parameter tuning p.30 3.6 Summary p.30 4 Model-based collaborative filtering based on CCAM algorithm p.32 4.1 Introduction p.32 4.2 Model-based collaborative filtering p.33 4.3 Experiments p.35 4.3.1 Data sets p.36 4.3.2 User feature selection p.36 4.3.3 Parameter Tuning p.37 4.3.4 Performance Comparison p.40 4.3.5 Application of tweet recommendation p.41 4.4 Summary p.42 5 Parallel co-clustering with augmented matrices algorithm for recommender system p.44 5.1 Introduction p.44 5.2 System flowchart p.46 5.2.1 Co-clustering with augmented matrices (CCAM) algorithm 47 5.2.2 Collaborative filtering based on CCAM p.48 5.3 Parallel co-clustering with augmented matrices (PCCAM) algorithm p.49 5.3.1 The PCCAM Algorithm for Input Format 1 p.50 5.3.2 The PCCAM Algorithm for Input Format 2 p.52 5.4 Model-based collaborative filtering algorithm p.52 5.4.1 Parallel collaborative filtering algorithm p.53 5.4.2 Hybrid method of discriminative model and generative model p.56 5.5 Experiments p.57 5.5.1 Data sets p.57 5.5.2 Efficiency of PCCAM algorithm p.58 5.5.2.1 Comparing PCCAM and CCAM algorithm p.58 5.5.2.2 Comparing execution-time between sparse vector and dense vector p.58 5.5.2.3 Effect of different factors p.59 5.5.2.4 Efficiency comparison of PCCAM and PSCOAL algorithm p.60 5.5.3 Effectiveness of PCCAM algorithm p.61 5.5.3.1 Parameter tuning p.62 5.5.3.2 Comparison with model-based collaborative filtering algorithms p.65 5.5.3.3 Effectiveness comparison of PCCAM and PSCOAL algorithm p.69 5.5.3.4 Sparsity p.69 5.5.3.5 Handling Cold-Start problem p.69 5.6 Summary p.70 6 Conclusion and Future Work p.72 Bibliography p.74 Appendix p.80 A Proof of Lemma 3.2.1 p.80 B Proof of Lemma 3.3.1 p.81 C Proof of Theorem 3.3.1 p.82

    [1] G. Adomavicius, and A. Tuzhilin (2005) Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions. IEEE Transactions on Knowledge and Data Engineering, pp. 734-749.
    [2] D. Agarwal and S. Merugu (2007) Predictive Discrete Latent Factor Models for Large Scale Dyadic Data. In KDD'07, Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 26-35.
    [3] R. Baeza-Yates and B. Ribeiro-Neto (1999) Modern Information Retrieval. Addison-Wesley, pp. 24-34.
    [4] A. Banerjee, I. Dhillon, J. Ghosh, S. Merugu, and D.S. Modha (2007) A generalized maximum entropy approach to Bregman co-clustering and matrix approximation. Journal of Machine Learning Research, pp. 1919-1986.
    [5] D. Billsus and M. J. Pazzani (1998) Learning collaborative information filters. In ICML '98, Proceedings of the 15th International Conference on Machine Learning, pp. 46-54.
    [6] G. Bisson and C. Grimal (2012) Co-clustering of multi-view datasets: a parallelizable approach. 2012 IEEE 12th International Conference on Data Mining (ICDM), pp. 828-833.
    [7] R. Bell and Y. Koren (2003) Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights. In ICDM'07, 7th IEEE International Conference on Data Mining, pp. 43-52.
    [8] D.-M. Blei, A.-Y. Ng, and M.-.I Jordan (2003) Latent Dirichlet allocation. Journal of Machine Learning Research, pp.993-1022.
    [9] B. Xu, J. Bu, C. Chen and D. Cai (2012) An Exploration of Improving Collaborative Recommender Systems via User-Item Subgroups. In WWW'12, Proceedings of the 21st international conference on World Wide Web, pp. 21-30.
    [10] G. Chen, F. Wang, and C. Zhang (2009) Collaborative filtering using orthogonal non-negative matrix tri-factorization. Information Processing and Management, Vol. 45, Issue 3, pp. 368-379.
    [11] T. Chen, L. Tang, Q. Liu, D. Yang, S. Xie, X. Cao, C. Wu, E. Yao, Z. Liu, Z. Jiang, C. Chen, W. Kong, Y. Yu (2012) Combining Factorization Model and Additive Forest for Collaborative Followee Recommendation. In KDD-Cup Workshop, 2012.
    [12] Y. Cheng and G. M. Church (2000) Biclustering of expression data. In ISMB'00, Proceedings of the 8th International Conference on Intelligent Systems for Molecular Biology, pp. 93-103.
    [13] H. Cho, I. Dhillon, Y. Guan, and S. Sra (2004) Minimum Sum-Squared Residue Co-Clustering of Gene Expression Data. In SDM'04, pp. 114-125.
    [14] T. Cover and J. Thomas (1991) Elements of Information Theory. John Wiley and Sons, ISBN:0-471-20061-1, pp. 12-49.
    [15] W. Dai, G.-R. Xue, Q. Yang, and Y. Yu (2007) Co-clustering based classification for out-of-domain documents. In KDD'07: Proceedings of the 13th ACM SIGKDD International conference on Knowledge Discovery and Data Mining, pp. 210-219.
    [16] S. Daruru, N. M. Marin, M. Walker and J. Ghosh (2009) Pervasive Parallelism in Data Mining: Dataflow Solution to Co-clustering Large and Sparse Netflix Data. In KDD'09, pp. 1115-1124.
    [17] J. Delgado and N. Ishii (1999) Memory-Based Weighted Majority Prediction for Recommender Systems. In 1999 SIGIR Workshop on Recommender Systems, University of California, Berkeley, pp. 1-5.
    [18] M. Deodhar and J. Ghosh (2007) A framework for simultaneous co-clustering and learning from complex data. In KDD'07, pp. 250-259.
    [19] M. Deodhar, C. Jones and J. Ghosh (2010) Parallel simultaneous co-clustering and learning with Map-Reduce. In IEEE International Conference on Granular Computing, pp. 149-154.
    [20] I.-S. Dhillon (2001) Co-clustering documents and words using bipartite spectral graph partitioning. In KDD'01: Proceedings of the 7th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 269-274.
    [21] S. Dhillon, S. Mallela, and D. S. Modha (2003) Information theoretic co-clustering. In KDD'03, Proceedings of the 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 89-98.
    [22] C. Ding, X. He, and H.-D. Simon (2005) On the equivalence of nonnegative matrix factorization and spectral clustering.} In Proceedings of the 5th SIAM international conference on data mining, pp. 606-610.
    [23] C. Ding, T. Li, W. Peng and H. Park (2006) Orthogonal nonnegative matrix tri-factorizations for clustering. In KDD'06: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 126-135.
    [24] P. Forbes and M. Zhu (2011) Content-boosted Matrix Factorization for Recommender Systems: Experiments with Recipe Recommendation. In RecSys'11, Proceedings of the 5th ACM conference on Recommender systems, pp. 261-264.
    [25] R. Gemulla, P. J. Haas, E. Nijkamp, and Y. Sismanis (2011) Large-scale matrix factorization with distributed stochastic gradient descent. In KDD'11, pp. 69-77.
    [26] T. George and S. Merugu (2005) A scalable collaborative filtering framework based on co-clustering. In ICDM '05, Proceedings of the 5th IEEE International Conference on Data Mining, pp. 625-628.
    [27] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann, and I. H. Witten (2009) The WEKA Data Mining Software: An Update. ACM SIGKDD Explorations Newsletter, Vol. 11, Issue 1, pp. 10-18.
    [28] J. Hannon, M. Bennett, and B. Smyth (2010) Recommending twitter users to follow using content and collaborative filtering approaches. Proceedings of the 4th ACM conference on Recommender systems, pp. 199-206.
    [29] A. Hartigan (1972) Direct Clustering of a Data Matrix. Journal of the American Statistical Association, Volume 67, Issue 337, pp. 123-129.
    [30] C. J. Hsieh and I. S. Dhillon (2011) Fast coordinate descent methods with variable selection for non-negative matrix factorization. In KDD'11, pp. 1064-1072.
    [31] W. Hill, L. Stead, M. Rosenstein, G. Furnas (1995) Recommending and Evaluating Choices in a Virtual Community of Use. In Proceedings of ACM CHI'95 Conference on Human Factors in Computing Systems, pp. 194-201.
    [32] Konstas, V. Stathopoulos, and J.-M. Jose (2009) On social networks and collaborative recommendation. In Proceedings of the 32nd international ACM SIGIR conference on Research and development, pp. 195-202.
    [33] D.-D. Lee and H.-S. Seung (1999) Learning the parts of objects by non-negative matrix factorization. Nature 401, pp. 788-791.
    [34] B. Li, Q. Yang, and X. Xue (2009) Can movies and books collaborate?: cross-domain collaborative filtering for sparsity reduction. In Proceedings of the 21st International Joint Conference on Artificial Intelligence, pp. 2052-2057.
    [35] B. Li, Q. Yang, and X. Xue. (2009) Transfer learning for collaborative filtering via a rating-matrix generative model. Proceedings of the 26th Annual International Conference on Machine Learning, pp. 617-624.
    [36] B. Long, Z. Zhang, and P.-S. Yu (2005) Co-clustering by Block Value Decomposition. In KDD'05, Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining, Chicago, Illinois, USA, ACM press, pp. 635-640.
    [37] P. Melville, R. J. Monney, and R. Nagarajan (2002) Content-boosted collaborative filtering for improved recommendation. In Proceedings of the 18th National Conference on Artificial Intelligence, pp. 187-192.
    [38] A. Mild and T. Reutterer (2001) Collaborative Filtering Methods for Binary Market Basket Data Analysis. In proceeding of: Active Media Technology, 6th International Computer Science Conference, pp. 302-313.
    [39] Narang, A. Srivastava, N.P.K. Katta (2012) High Performance Offline and Online Distributed Collaborative Filtering. In ICDM'12, pp. 549-558.
    [40] P. S. Pacheco (1996) Parallel Programming with MPI. Morgan Kaufmann Publishers, ISBN 9781558603394, pp. 1-500.
    [41] S. Papadimitriou and J. Sun (2008) Disco: distributed co-clustering with Map-Reduce. In ICDM'08, 8th IEEE International Conference on Data Mining, pp. 512--521.
    [42] W. Pan and Q. Yang (2013) Transfer learning in heterogeneous collaborative filtering domains. Journal of Artificial Intelligence, Vol. 197, pp. 39-55.
    [43] V. Ramanathan (2010) Parallelizing an Information Theoretic Co-clustering Algorithm Using a Cloud Middleware. International Conference on Data Mining Workshops (ICDMW), pp. 186-193.
    [44] Sarwar, G. Karypis, J. Konstan, and J. Riedl (2000) Application of dimensionality reduction in recommender systems: a case study. In WebKDD-2000 Workshop.
    [45] Sarwar, G. Karypis, J. Konstan, and J. Riedl (2001) Item-based collaborative filtering recommendation algorithms. Proceedings of the 10th international conference on World Wide Web, ACM, pp. 285-295.
    [46] J.-B. Schafer, D. Frankowski, J. Herlocker, and S. Sen (2007) Collaborative filtering recommender systems. In The Adaptive Web, pp. 291-324.
    [47] M. Shafiei and E. Milios (2005) Model-based Overlapping Co-Clustering. In KDD'05, Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining, pp. 532-537.
    [48] M. Shafiei and E. Milios (2006) Latent Dirichlet Co-Clustering. In ICDM'06, the 6th IEEE International Conference on Data Mining, pp. 542-551.
    [49] H. Shan and A. Banerjee (2008) Bayesian Co-clustering. In ICDM'08, the 8th IEEE International Conference on Data Mining 2008, pp. 530-539.
    [50] Shi and L. Li (2012) High performance genetic algorithm based text clustering using parts of speech and outlier elimination. Journal of Applied Intelligence, Vol. 38, Issue 4, pp. 511-519.
    [51] N. Slonim and N. Tishby (2000) Document clustering using word clusters via the information bottleneck method. In proceedings of the 23rd annual international ACM SIGIR conference on research and development in information retrieval. Athens, Greece, pp. 208-215.
    [52] Sugiyama, K. Hatano, and M. Yoshikawa (2004) Adaptive web search based on user profile constructed without any effort from users. International World Wide Web Conference Proceedings of the 13th international conference on World Wide Web, pp. 675-684.
    [53] W. Scott (2009) Sturges' rule. WIREs Computational Statistics, pp. 303-306.
    [54] E. Wall, A. Rechtsteiner, and L. M. Rocha (2003) Singular value decomposition and principal component analysis. In A Practical Approach to Microarray Data Analysis. D.P. Berrar, W. Dubitzky, M. Granzow, eds. Kluwer: Norwell, MA. LANL LA-UR-02-4001, pp. 91-109.
    [55] L. Wu, C. H. Chang, and R. Z. Liu (2011) Collaborative Filtering with CCAM. 10th International Conference on Machine Learning and Applications and Workshops, Vol. 2, pp. 245-250.
    [56] M.-L. Wu, C.-H. Chang, and R.-Z. Liu (2013) Co-clustering with augmented matrix. Journal of Applied Intelligence, Vol. 39, Issue 1, pp. 153-164.

    [57] M.-L. Wu, C.-H. Chang, and R.-Z. Liu (2014) Integrating content-based filtering with collaborative filtering using co-clustering with augmented matrices. Expert Systems with Applications, Vol. 41, Issue 6, pp. 2754-2761.
    [58] B. Xu, J. Bu, C. Chen, and D. Cai (2012) An Exploration of Improving Collaborative Recommender Systems via User-Item Subgroups. Proceedings of the 21st international conference on World Wide Web, pp. 21-30.
    [59] H. F. Yu, C. J. Hsieh, S. Si, and I. Dhillon (2012) Scalable coordinate descent approaches to parallel matrix factorization for recommender systems. In ICDM'12, Proceedings of the 2012 IEEE 12th International Conference on Data Mining, pp. 765-774.
    [60] W. Zhao, H. Ma, and Q. He (2009) Parallel K-Means Clustering Based on MapReduce. In CloudCom 2009, pp. 674-679.
    [61] Y. Zhou, D. Wilkinson, R. Schreiber, and R. Pan (2008) Large-scale parallel collaborative filtering for the Netflix prize. In Proceedings of the 4th international conference on Algorithmic Aspects in Information and Management, pp. 337-348.

    QR CODE
    :::