電子商務的興起、多媒體的技術成熟，網路上資訊的數量和種類也隨之增加，但大量的資訊對於使用者來說並不完全是件好事，絕大部分資訊對特定使用者來說是無關的，也就是資訊超載問題。本研究探討資訊超載的對象為多媒體影片，多媒體不僅是知識傳播的媒介，在人類的娛樂生活扮演著重要的角色，然而多媒體的數量之龐大，要從中挑選想看的是何其費力；個人化多媒體推薦可幫助使用者篩選出使用者想看的多媒體，替使用者省下自行搜尋的時間。
大眾分類法是Web 2.0網站的熱門應用，系統允許使用者在項目標記標籤來代表該項目的分類或特色，而這些標籤資訊可能直接或間接地反映出使用者的個人興趣，在推薦技術中加入標籤資訊能提高推薦績效，故將發展標籤協同過濾技術。
除了標籤協同過濾技術，情境感知的推薦也是本研究想探討的議題。由於科技的進步，電子行動裝置的功能越來越強大，播放多媒體也成為基本的規格，也因此使用者觀看多媒體已不再拘限於家中。然而行動裝置本身在規格上有所限制，在進行多媒體推薦理當要考慮到這點；使用者可在各種場合觀賞多媒體，不同場合適合觀賞的多媒體也不同。
除了在改良推薦技術的績效和考量情境因子外，本研究實作個人化推薦系統，幫助使用者過濾大量的多媒體，使用者透過網路登入到推薦系統，系統便會推薦滿足使用者興趣且符合使用者情境的多媒體列表。

Because electronic commerce has been flourishing in recent year, the amount and the variety of information on the web have also been rapidly increasing. However, many problems occur as the result of information overload. This thesis is to study the issue of information overload in the field of multimedia that covers not only medium of diffuse knowledge but also entertainment of everyday life. The main goal of this work is to use personalized recommendation technologies to help users select multimedia he is interested in.
The thesis investigates two types of personalized recommendation: tag-based recommendation and context-aware recommendation. Regarding the former kind of recommendation, Folksonomy is the popular Web2.0 application that allows users tagging items to indicate the corresponding characteristics. These tags, provided by the users, directly or indirectly reflect his personal interests. Therefore the recommendation performance is enhanced when the tags are used with computational methods. On the other hand, the latter kind focuses on the contents and the relevant situations, because what multimedia is considered suitable for users can be different under different situations. The advantages of the personalized recommendation technology can improve performance of recommendation and take the context into account at the same time. Meanwhile this study also implements a working system for personalized multimedia recommendation.

第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究目的 3
1.3 研究流程 4
第二章 文獻探討 5
2.1 推薦技術 5
2.1.1 內容基礎推薦技術 5
2.1.2 協同過濾推薦技術 5
2.1.3 其他推薦推薦技術 6
2.1.4 推薦技術相關研究 7
2.2 標記標籤與大眾分類法 7
2.3 情境 11
第三章 研究架構 14
第四章 推薦技術演算法與情境規則 18
4.1 推薦技術演算法 18
4.1.1 Decision Tree 18
4.1.2 Support Vector Machine 20
4.1.3 Naive Bayes 22
4.1.4 Collaborative Filtering 22
4.1.5 Tag-based Collaborative Filtering 23
4.2 情境感知規則 27
4.2.1. 通訊條件 28
4.2.2. 存取設備 29
4.2.3. 社會互動 30
4.2.4. 地點 31
第五章 實驗與結果 33
5.1 實驗數據來源 33
5.1.1 使用者評價 33
5.1.2 多媒體元資料 33
5.2 評估準則（Evaluation criterion） 34
5.3 實驗 36
5.3.1 實驗一 Tag-CF與CF在不同鄰居相似度門檻值其績效的比較 37
5.3.2 實驗二 Tag-CB在三種分類演算法之績效比較 39
5.3.3 實驗三 Keyword-CB在三種分類演算法之績效比較 41
5.3.4 實驗四 Tag-CF、Tag-CB SVM和Keyword-CB SVM之績效比較 42
第六章 推薦系統實作 44
6.1 系統簡介 44
6.2 系統之情境感知 44
6-3 系統功能 47
6.4 系統畫面展示 48
第七章 結論與未來研究 53
7.1 結論 53
7.2 未來研究 53
參考文獻 55

[1] Retrieved from The Internet Movie Database: http://www.imdb.com/
[2] Retrieved from Delicious: http://delicious.com/
[3] Retrieved 2008, from Java Movie Database: http://www.jmdb.de/
[4] Retrieved 2008, from YouTube: http://www.youtube.com/
[5] Belkin, N. J., & Croft, W. B. (1992). Information filtering and information retrieval: two sides of the same coin? Communications of the ACM, Volume 35, Issue 12 , pp. 29-38.
[6] Breese, J. S., Heckerman, D., & Kadie, C. (1998). Empirical Analysis of Predictive Algorithms for Collaborative Filtering. Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence., (pp. 43-52).
[7] Burke, R. (2002). Hybrid Recommender Systems: Survey and Experiments. User Modeling and User-Adapted Interaction, Volume 12, Issue 4 , pp. 331-370 .
[8] Cheung, K.-W., & Tian, L. F. (2004). Learning User Similarity and Rating Style for Collaborative Recommendation. Information Retrieval, Volume 7, Issue 3-4 , pp. 395-410.
[9] Degemmis, M., Lops, P., & Semeraro, G. (2007). A content-collaborative recommender that exploits WordNet-based user profiles for neighborhood formation. User Modeling and User-Adapted Interaction, Volume 17, Issue 3 , pp. 217-255.
[10] Goldberg, D., Nichols, D., Oki, B. M., & Terry, D. (1992). Using collaborative filtering to weave an information tapestry. Communications of the ACM , pp. 61-70.
[11] Golder, S. A., & Huberman, B. A. (2005). The Structure of Collaborative Tagging Systems. Information Dynamics Lab, HP Labs.
[12] Hand, D. J., Mannila, H., & Smyth, P. (2001). Principles of Data Mining. The MIT Press.
[13] Katz, E., Blumler, J., & Gurevitch, M. (1975). The Uses of Mass Communications: Current Perspectives on Gratifications Research. Sage Publications, Inc.
[14] Lee, J. S., & Lee, J. C. (2007). Context Awareness by Case-Based Reasoning in a Music Recommendation System. In Ubiquitous Computing Systems (pp. 45-58). Springer Berlin.
[15] Lekakos, G., & Caravelas, P. (2006). A hybrid approach for movie recommendation. Multimedia Tools and Applications, Volume 36, Numbers 1-2 , pp. 55-70.
[16] Marlow, C., Naaman, M., Boyd, D., & Davis, M. (2006). HT06, tagging paper, taxonomy, Flickr, academic article, to read. Proceedings of the seventeenth conference on Hypertext and hypermedia (pp. 31-40). ACM.
[17] Motion picture rating system. (2008). Retrieved 6 2009, from Wikipedia: http://en.wikipedia.org/wiki/Motion_picture_rating_system
[18] Ozturk, P. (1998). A context-sensitive, iterative approach to diagnostic problem solving. In Methodology and Tools in Knowledge-Based Systems (pp. 546-557).
[19] Pogacnik, M., Tasic, J., Meza, M., & Kosir, A. (2005, 11). Personal Content Recommender Based on a Hierarchical User Model for the Selection of TV Programmes. User Modeling and User-Adapted Interaction, Volume 15, Number 5 , pp. 425-457.
[20] Pu, P., & Chen, L. (2006). Trust building with explanation interfaces. International Conference on Intelligent User Interfaces (pp. 93-100). ACM.
[21] Quinlan, J. R. (1992). C4.5: programs for machine learning. Morgan Kaufmann Publishers Inc. San Francisco, CA, USA.
[22] Rashid, A. M., Albert, I., Cosley, D., Lam, S. K., McNee, S. M., Konstan, J. A., et al. (2002). Getting to know you: learning new user preferences in recommender systems. International Conference on Intelligent User Interfaces, (pp. 127-134).
[23] Riecken, D. (2000). Introduction: personalized views of personalization. Communications of the ACM, Volume 43, Issue 8 , pp. 26-28.
[24] Schilit, B., Adams, N., & Want, R. (1994). Context-Aware Computing Applications. IEEE Workshop on Mobile Computing Systems and Applications, (pp. 89-101).
[25] Schmidt, A., Beigl, M., & Gellersen, H.-w. (1999). There is more to Context than Location. Computers and Graphics , pp. 893-901.
[26] Setten, M. v., Pokraev, S., & Koolwaaij, J. (2004). Context-Aware Recommendations in the Mobile Tourist Application COMPASS. In Adaptive Hypermedia and Adaptive Web-Based Systems (pp. 235-244). Springer Berlin.
[27] Shardanand, U., & Maes, P. (1995). Social Information Filtering: Algorithms for Automating "Word of Mouth". Proceedings of the SIGCHI conference on Human factors in computing systems (pp. 210-217). ACM.
[28] Vapnik, V. (1995). The Nature of Statistical Learning Theory. Springer.
[29] Wal, T. V. (2007). Folksonomy Coinage and Definition. Retrieved 6 29, 2009, from vanderwal.net: http://vanderwal.net/folksonomy.html