跳到主要內容

臺灣博碩士論文加值系統

(3.236.28.137) 您好!臺灣時間:2021/07/25 21:16
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:陳映霖
研究生(外文):Ying-Lin Chen
論文名稱:結合時間相依性之推薦系統
論文名稱(外文):Incorporating Temporal Dependencies in Recommendation Systems
指導教授:鄧維光
指導教授(外文):Wei-Guang Teng
學位類別:碩士
校院名稱:國立成功大學
系所名稱:工程科學系專班
學門:工程學門
學類:綜合工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:英文
論文頁數:39
中文關鍵詞:推薦系統協同式過濾時間相依性生命週期時間距離使用期限
外文關鍵詞:life cycletemporal dependenciescollaborative filteringrecommendation systemtime distancelifespan
相關次數:
  • 被引用被引用:0
  • 點閱點閱:485
  • 評分評分:
  • 下載下載:121
  • 收藏至我的研究室書目清單書目收藏:1
為了幫助使用者找到其可能喜好的物品,推薦系統大量利用了資料分析的技巧,其中使用者評價資料與相關的運用機制在推薦系統相關研究工作中扮演了很重要的角色,而在眾多的推薦技術當中,協同式過濾是最廣泛使用且最受認同的方法,其能正確地估計使用者對諸多未曾看過物品的評價,並將預估具有最高評價的物品推薦給使用者;在實際生活中,這樣的推薦系統可應用於推薦餐廳、旅遊景點、電影、書籍及音樂光碟等。在本論文中,我們在推薦系統中加入了時間相依性的觀念,亦即使用者的喜好會隨著時間而變化,因此在採用協同式過濾方法來進行推薦時,也會根據時間因素來尋求與使用者喜好類似的同儕意見;明確地而言,藉由引入評價資料具有其生命週期的觀念,我們定義了 "時間距離" 與 "使用期限" 這兩個參數,其中兩筆資料間的時間距離可經由生命週期之定義所推算得出,而使用期限則反映了資料新鮮度的衰減速度。整體而言,我們在本論文中提出了一可通用的公式以更進一步地增進協同式過濾的推薦準確度,同時我們也採用了可公開取得之電影評價資料庫以進行評估實驗,經過實證研究後可發現我們的方法可有效地降低推薦系統的誤差率。
Recommendation systems are usually capable of utilizing data analysis techniques to help users find the items they would like. The rating structure and corresponding estimation mechanisms play a critical role in recommendation system studies. Among many alternatives, the collaborative filtering technique is generally accepted to be the most successful one in estimating user ratings of unseen items and thus to derive appropriate recommendations for users. Example applications include the recommendation systems for restaurants, tours, movies, books, CDs and many others. In this work, the concept of temporal dependencies is introduced. In other words, the interest of a user changes with time. Consequently, when utilizing the collaborative filtering to offer recommendations to a user, peers of similar tastes can be identified by considering the temporal dependencies. Specifically, two parameters are explicitly defined, i.e., time distance, and lifespan, by considering the life cycle embedded in rating records. The time distance between two data records can be calculated by referring the length of a life cycle. Also, lifespan reflects the decay rate of data freshness as time advances. Generally speaking, we propose in this work a general formula to further improve the recommendation precision of the collaborative filtering technique. Moreover, public datasets of movie ratings are adopted as our testbed for evaluation. Empirical studies show that our approach can effectively reduce the estimation error as compared with a typical recommendation system.
Chapter 1 Introduction 1
1.1 Motivation and Overview of the Thesis 1
1.2 Contributions of the Thesis 2
Chapter 2 Preliminaries 4
2.1 Overview of Recommendation Techniques 4
2.2 Collaborative Filtering Techniques 5
2.3 Temporal Dependencies and Corresponding Usage in Recommendation Systems 8
Chapter 3 Utilization of Temporal Dependencies in Recommendation Systems 11
3.1 An Illustrative Examples of CF Recommendation 11
3.1.1 Rating Estimation with User-based CF Technique 11
3.1.2 Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) 13
3.2 Time Weight Algorithms for Temporal Dependencies 15
3.2.1 Time Decay Functions 15
3.2.2 Absolute Time, Time Difference, Life Cycle and Time Distance 16
3.2.3 Time Function with Time Distance and Lifespan 18
3.2.4 Time Weight Algorithm with Time Function 20
3.3 Test Result of Time Weight Algorithms 21
3.3.1 Testing Results on a Small Dataset 21
3.3.2 Testing Results on More Sampled Datasets 23
Chapter 4 Empirical Studies 25
4.1 Experimental Environment 25
4.2 Experimental Results on the MovieLens Dataset 26
4.2.1 Impact of kNN 26
4.2.2 Impact of Lifespan 27
4.3 Comparison with Prior Studies 30
Chapter 5 Conclusions and Future Works 34
Bibliography 35
[1]G. Adomavicius, R. Sankaranarayanan, S. Sen, and A. Tuzhilin,“Incorporating Contextual Information in Recommender Systems Using a Multidimensional Approach,” ACM Transactions on Information Systems, 23(1):103-145, January 2005.
[2]G. Adomavicius and A. Tuzhilin, “Multidimensional Recommender Systems: A Data Warehousing Approach,” Proceedings of the Second International Workshop on Electronic Commerce, pages 180-192, November 2001.
[3]K. Ali and W. V. Stam, “TiVo: Making Show Recommendations Using a Distributed Collaborative Filtering Architecture,” Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 394-401, August 2004.
[4]R. M. Bell and Y. Koren, “Scalable Collaborative Filtering with Jointly Derived Neighborhood Interpolation Weights,” Proceedings of the 7th IEEE International Conference on Data Mining, pages 43-52, October 2007.
[5]R. M. Bell, Y. Koren and C. Volinsky, “Modeling Relationships at Multiple Scales to Improve Accuracy of Large Recommender Systems,” Proceedings of the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 95-104, August 2007.
[6]Book-Crossing Dataset, http://www.informatik.uni-freiburg.de/~cziegler/BX/.
[7]M. Deshpande and G. Karypis, “Item-based Top-N Recommendation Algorithms,” ACM Transactions on Information Systems, 22(1):143-177, January 2004.
[8]Y. Ding and X. Li, “Time Weight Collaborative Filtering,” Proceedings of the 14th ACM International Conference on Information and Knowledge Management, pages 95-104, October 2005.
[9]Y. Ding, X. Li, and M. E. Orlowska, “Recency-Based Collaborative Filtering,” Proceedings of the 17th Australasian Database Conference, 99-107, January 2006.
[10]W. Fan, “Systematic Data Selection to Mine Concept-drifting Data Streams,” Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 128 -137, August 2004.
[11]D. Goldberg, D. Nichols, B. M. Oki and D. Terry, “Using Collaborative Filtering to Weave an Information Tapestry,” Communications of the ACM, 35(12):61-70, December 1992.
[12]GroupLens Research, http://www.grouplens.org/.
[13]J. L. Herlocker and J. A. Konstan, “Content-Independent Task-Focused Recommendation,” IEEE Internet Computing, 5(6):40-47, November 2001.
[14]J. L. Herlocker, J. A. Konstan, A. Borchers and John Riedl, “An Algorithmic Framework for Performing Collaborative Filtering,” Proceedings of the 22nd ACM SIGIR Conference on Information Retrieval, pages 230-237, August 1999.
[15]Jester Joke Dataset, http://www.ieor.berkeley.edu/~goldberg/jester-data/.
[16]J. Konstan, B. Miller, D. Maltz, J. Herlocker, L. Gordon and J. Riedl, “GroupLens: Applying Collaborative Filtering to UsenetNews,” Communications of the ACM, 40(3):77-87, March 1997.
[17]T. Q. Lee, Y. Park, and Y.-T. Park, “A Time-based Approach to Effective Recommender Systems Using Implicit Feedback,” Expert Systems with Applications, 34(4):3055-3062, May 2008.
[18]G. Linden, B. Smith and J. York, “Amazon.com Recommendations: Item-to-item Collaborative Filtering,” IEEE Internet Computing, 7(1):76-80, January 2003.
[19]MovieLens, http://www.movielens.org/.
[20]P. Resnick, N. Iacovou, M. Suchak, P. Bergstorm and J. Riedl, “GroupLens: An Open Architecture for Collaborative Filtering of Netnews,” Proceedings of the ACM Conference on Computer Supported Cooperative Work, pages 175-186, October 1994.
[21]B. Sarwar, G. Karypis, J. Konstan and J. Reidl, “Analysis of Recommendation Algorithms for E-commerce,” Proceedings of the 2nd ACM Conference on Electronic Commerce, pages 158-167, October 2000.
[22]B. Sarwar, G. Karypis, J. Konstan and J. Reidl, “Item-based Collaborative Filtering Recommendation Algorithms,” Proceedings of the 10th International Conference on World Wide Web, pages 285-295, May 2001.
[23]K. Sugiyama, K. Hatano, and M. Yoshikawa, “Adaptive Web Search Based on User Profile Constructed without Any Effort from Users,” Proceedings of the 13th International Conference on World Wide Web, pages 675-684, May 2004.
[24]T. Y. Tang, P. Winoto, and K. C. C. Chan, “On the Temporal Analysis for Improved Hybrid Recommendations,” Proceedings of the 2003 IEEE/WIC International Conference on Web Intelligence, pages 214-220, October 2003.
[25]L. Terveen, J. McMackin, B. Amento, and W. Hill, ”Specifying Preferences Based on User History,” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pages 315-322, April 2002.
[26]VoteWorld, http://ucdata.berkeley.edu:7101/new_web/VoteWorld/voteworld/datasets.html.
[27]H. Wang, W. Fan, P. S. Yu, and J. Han, “Mining Concept-drifting Data Streams Using Ensemble Classifiers,” Proceedings of the 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 226-235, August 2003.
[28]Yahoo! Movies, http://movies.yahoo.com/.
[29]Y. Zhao, C. Zhang, and S. Zhang, “A Recent-Biased Dimension Reduction Technique for Time Series Data,” Advances in Knowledge Discovery and Data Mining, 3518:751-757, May 2005.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊