隨著資通訊技術與行動網路快速的演進，使得我們能夠取得的多媒體資訊管道極其豐富。然後伴隨著資料量以超越等比級數的方式快速成長，在大量的多媒體資訊下，使用者必須花費驚人的搜尋時間才能取得自身所喜好的多媒體項目，從此欣賞多媒體不再是樂事，而是一項惱人且費時的工作。
有鑑於此，本研究提出了以協同過濾法當中的最近鄰居法以及矩陣分解方法，各別結合使用者對於推薦項目的評分以及多個情境資訊的多種方法，透過機器學習的方式來有效的過濾大量的多媒體資訊，幫助使用者用比以往更有效率、更快速的取得自身所感興趣的項目成為了本研究最主要的貢獻。而透過最後的數據呈現以及完整與現有文獻的對照，可以知道本研究的實驗結果無論是在平均的推薦績效以及穩定性上皆優於傳統的作法。

The rapid development of mobile device and multimedia allows us today to have extremely rich multimedia information sources. The numbers of data also increase rapidly so that users have to spend more and more time to find their preferred multimedia items among a multitude of information.
To tackle this problem, the collaborative filtering (CF) approach to recommender systems has recently been proposed and used. The present study develops some techniques including a machine learning approach for performance improvement. The aim is to effectively filter large amounts of multimedia information and to achieve competitive accuracy by incorporating multi-contexts into the neighborhood and matrix factorization methods. Experiments have been conducted and the results show that the proposed approach outperforms other context-aware recommendation methods.

論文審定書 i
摘要 ii
Abstract iii
第一章 緒論 1
1.1 背景與動機 1
1.2 研究目的 2
1.2.1 考量多個情境因素推薦 3
1.2.2 建構CF推薦系統 3
1.2.3 推薦技術績效比較 3
1.3 研究流程 4
第二章 文獻探討 6
2.1 推薦技術 6
2.1.1 內容基礎推薦技術 6
2.1.2 協同過濾推薦技術 7
2.2 情境感知 11
2.3 協同過濾法結合情境因子 12
第三章 研究方法與步驟 14
3.1 研究系統架構 14
3.1.1 Multimedia Dataset 14
3.1.2 Data Preprocessing 15
3.1.3 Collaborative Filtering Techniques 15
3.1.4 Recommendation System 15
3.2 資料集介紹與處理 16
3.2.1 LDOS-CoMoDa Dataset 17
3.2.2 MovieLens Dataset 19
3.2.3 Sushi Preference Dataset 20
3.2.4 AIST Food Dataset 21
3.3 標準協同過濾方法實作 22
3.3.1 KNN標準系統實作 22
3.3.2 SVD標準系統實作 26
3.4 結合多情境之協同過濾法 31
3.4.1 KNN－多維度向量方法 32
3.4.2 KNN－循序式方法 33
3.4.3 KNN－壓縮式輔助方法 37
3.4.4 SVD－Baseline +方法 38
3.4.5 SVD－精煉式方法 42
第四章 實驗與結果 43
4.1 方法評量與績效 43
4.1.1 KNN－多維度向量方法 (MDKNN) 43
4.1.2 KNN－循序式方法 (SKNN) 46
4.1.3 KNN－壓縮式方法 (CKNN) 50
4.1.4 KNN－多維度向量+壓縮法 (C+MDKNN) 51
4.1.5 KNN－循序+壓縮法 (C+SKNN) 52
4.1.6 KNN－多維度向量+循序+壓縮法 (C+S+MDKNN) 56
4.1.7 SVD－Baseline Plus方法 (Baseline+) 61
4.1.8 SVD－精煉式方法 (RSVD) 62
4.2 研究比較與分析 65
4.3 模擬個案 72
第五章 研究貢獻與未來展望 79
5.1 研究貢獻 79
5.2 未來展望 80
參考文獻 82

[1] Adomavicius, Gediminas, and Alexander Tuzhilin. "Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions." Knowledge and Data Engineering, IEEE Transactions on 17.6 (2005): 734-749.
[2] Adomavicius, Gediminas, and YoungOk Kwon. "Improving aggregate recommendation diversity using ranking-based techniques." Knowledge and Data Engineering, IEEE Transactions on 24.5 (2012): 896-911.
[3] Belkin, Nicholas J., and W. Bruce Croft. "Information filtering and information retrieval: two sides of the same coin?." Communications of the ACM 35.12 (1992): 29-38.
[4] Chen, Liming, and Chris Nugent. "Ontology-based activity recognition in intelligent pervasive environments." International Journal of Web Information Systems 5.4 (2009): 410-430.
[5] Chou, Shih-Chun, et al. "Semantic web technologies for context-aware museum tour guide applications." Advanced Information Networking and Applications, 2005. AINA 2005. 19th International Conference on. Vol. 2. IEEE, 2005.
[6] Dey, Anind K. "Understanding and using context." Personal and ubiquitous computing 5.1 (2001): 4-7.
[7] Google Play, Wikipedia, http://zh.wikipedia.org/wiki/Google_Play, May 2014, [Retrieved 2014/05].
[8] Goldberg, David, et al. "Using collaborative filtering to weave an information tapestry." Communications of the ACM 35.12 (1992): 61-70.
[9] Guan, Linting, and Hailing Lu. "Recommend Items for User in Social Networking Services with CF." Computer Science & Service System (CSSS), 2012 International Conference on. IEEE, 2012.
[10] He, Zhenyu, and Lianwen Jin. "Activity recognition from acceleration data based on discrete consine transform and SVM." Systems, Man and Cybernetics, 2009. SMC 2009. IEEE International Conference on. IEEE, 2009.
[11] Intille, Stephen S., et al. "A living laboratory for the design and evaluation of ubiquitous computing technologies." CHI'05 extended abstracts on Human factors in computing systems. ACM, 2005.
[12] Joe, Konstan, Grouplens.org, http://grouplens.org/datasets/movielens/, January 2014,[Retrieved 2014/06].
[13] Kantor, Paul B., et al. Recommender systems handbook. Springer, 2011.
[14] Koren, Yehuda, and Robert Bell. "Advances in collaborative filtering."Recommender Systems Handbook. Springer US, 2011. 145-186.
[15] Koren, Yehuda, Google Scholar, http://scholar.google.com/citations?user=wTmI_ HYAAAAJ, June 2014, [Retrieved 2014/06].
[16] Koren, Yehuda. "Factorization meets the neighborhood: a multifaceted collaborative filtering model." Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2008.
[17] Koren, Yehuda, Robert Bell, and Chris Volinsky. "Matrix factorization techniques for recommender systems." Computer 42.8 (2009): 30-37.
[18] LDOS-CoMoDa dataset, University of Ljubljana, July 2012, http://212.235.18 7.145/spletnastran/raziskave/um/comoda/comoda.php.
[19] Lee, Jae Sik, and Jin Chun Lee. "Context awareness by case-based reasoning in a music recommendation system." Ubiquitous Computing Systems. Springer Berlin Heidelberg, 2007. 45-58.
[20] Li, Qikai, Shijian Li, and Gang Pan. "ScudMobile: A context-aware information service system for car touring in city-wide area." Intelligent Vehicles Symposium, 2009 IEEE. IEEE, 2009.
[21] Meehan, Kevin, et al. "Context-aware intelligent recommendation system for tourism." Pervasive Computing and Communications Workshops (PERCOM Workshops), 2013 IEEE International Conference on. IEEE, 2013.
[22] Mobasher, Bamshad, Xin Jin, and Yanzan Zhou. "Semantically enhanced collaborative filtering on the web." Web Mining: From Web to Semantic Web. Springer Berlin Heidelberg, 2004. 57-76.
[23] Nguyen, Trung V., Alexandros Karatzoglou, and Linas Baltrunas. "Gaussian Process Factorization Machines for Context-aware Recommendations."
[24] Park, Heum, Soonho Kwon, and Hyuk-Chul Kwon. "Ontology-based approach to intelligent ubiquitous tourist information system." Ubiquitous Information Technologies & Applications, 2009. ICUT'09. Proceedings of the 4th International Conference on. IEEE, 2009.
[25] Riboni, Daniele, et al. "Is ontology-based activity recognition really effective?."Pervasive Computing and Communications Workshops (PERCOM Workshops), 2011 IEEE International Conference on. IEEE, 2011.
[26] Riboni, Daniele, and Claudio Bettini. "COSAR: hybrid reasoning for context-aware activity recognition." Personal and Ubiquitous Computing 15.3 (2011): 271-289.
[27] Sarwar, Badrul, et al. "Item-based collaborative filtering recommendation algorithms." Proceedings of the 10th international conference on World Wide Web. ACM, 2001.
[28] Su, Xiaoyuan, and Taghi M. Khoshgoftaar. "A survey of collaborative filtering techniques." Advances in artificial intelligence 2009 (2009): 4.
[29] Sarwar, Badrul, et al. Application of dimensionality reduction in recommender system-a case study. No. TR-00-043. Minnesota Univ Minneapolis Dept of Computer Science, 2000.
[30] Schilit, Bill, Norman Adams, and Roy Want. "Context-aware computing applications." Mobile Computing Systems and Applications, 1994. WMCSA 1994. First Workshop on. IEEE, 1994.
[31] Van Setten, Mark, Stanislav Pokraev, and Johan Koolwaaij. "Context-aware recommendations in the mobile tourist application COMPASS." Adaptive hypermedia and adaptive web-based systems. Springer Berlin Heidelberg, 2004.
[32] Van Kasteren, Tim, et al. "Accurate activity recognition in a home setting."Proceedings of the 10th international conference on Ubiquitous computing. ACM, 2008.
[33] Wu, Ka Cheuk, and Dickson KW Chiu. "Toward Tourist Service Integration and Personalization with Semantic Web Services: A Case Study in Hong Kong." e-Business Engineering, 2008. ICEBE'08. IEEE International Conference on. IEEE, 2008.
[34] Waga, Karol, Andrei Tabarcea, and Pasi Fränti. "Context aware recommendation of location-based data." 2011 15th International Conference on System Theory, Control, and Computing (ICSTCC). 2011.
[35] Wu, Mingrui. "Collaborative filtering via ensembles of matrix factorizations."Proceedings of KDD Cup and Workshop. Vol. 2007. 2007.
[36] Xiao, Hua, et al. "An approach for context-aware service discovery and recommendation." Web Services (ICWS), 2010 IEEE International Conference on. IEEE, 2010.
[37] Yueh, Yves TF, et al. "A virtual travel agent system for m-tourism with semantic web service based design and implementation." Advanced Information Networking and Applications, 2007. AINA'07. 21st International Conference on. IEEE, 2007.
[38] Youtube press, Youtube, https://www.youtube.com/yt/press/zh-TW/, May 2013, [Retrieved 2014/05].
[39] Cui, Yongli, et al. "A Collaborative Filtering Algorithm Based on User Activity Level." Business Intelligence and Financial Engineering (BIFE), 2012 Fifth International Conference on. IEEE, 2012.
[40] Dao, Tuan Hung, Seung Ryul Jeong, and Hyunchul Ahn. "A novel recommendation model of location-based advertising: Context-Aware Collaborative Filtering using GA approach." Expert Systems with Applications39.3 (2012): 3731-3739.
[41] Rendle, Steffen. "Factorization machines." Data Mining (ICDM), 2010 IEEE 10th International Conference on. IEEE, 2010.
[42] Ye, Mao, Peifeng Yin, and Wang-Chien Lee. "Location recommendation for location-based social networks." Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems. ACM, 2010.