臺灣博碩士論文加值系統

由於網際網路的普及化，使得網路的服務與經營型態層出不窮，而在Web 2.0興起後，促成使用者們積極在網站上交流、互動，資訊分享與個人化管理更成為網路服務的主流。社會性標籤是Web 2.0的一項重要服務，社會性標籤讓網路使用者對網路資源進行標記，實現了運用大眾的力量管理及分享網路資源的機制。社會性標籤之所以能夠成為當前的主流，原因是其自由定義標籤與分享個人知識，讓大眾不受侷限地定義個人化標籤，且能夠透過標籤產生互動。
傳統基於受歡迎程度的標籤推薦方法能夠根據當前在網路上其他使用者對欲推薦之資源標記的標籤頻率，推薦目標使用者最多人拿來標記的標籤。然而，這類型的推薦方法並未考量到個人化的部分，因此推薦結果不一定適合目標使用者。本研究將使用者本身的標籤使用經驗與其他使用者標記欲推薦之資源的標籤使用經驗一同納入考量，採用一種序列模式的方式來計算出標籤權重。最後，將近年許多研究都在探討的時間特性也一併作為標籤權重的變因。透過使用者在不同時間點上有不同的標籤使用習慣來預測標籤，達成標籤個人化推薦的目的。

There are many kinds of service on internet because of the popularize of internet. And the rise of Web 2.0 stimulates users interact on internet. Information sharing and personalize management become the main web service. The social tagging system is an important application of Web 2.0. The social tagging system allows users tag internet resources. This implement the mechanism of internet resource sharing and management by the power of population. The social tagging system can become the main application is because it can freely define tags and sharing personal knowledges. Everyone can define personalized tags without restrictions and interact by tags.
In the past tag recommendation is based on the frequency that resources had been tagged. And the tags that had tagged the most would be recommend to users. However, this way do not consider the personalization. The recommendation may not be what users need. This study both considers user’s own tag usage experiences and other users’ tag usage experiences, which they had tagged to resource. Using the sequence pattern way to calculate the weight of tags. Our study also consider the temporal characteristic, which recent researches are focus on. Predicting tags by users’ different tagging actions at different time to achieve the goal of recommend tags.

摘要 I
Abstract II
目錄 III
表目錄 V
圖目錄 VI
第一章、 緒論 1
第一節、 研究背景 1
第二節、 動機與目的 2
第三節、 論文架構 4
第二章、 文獻探討 5
第一節、 社會性標籤 5
第二節、 基於社會性標籤下的標籤推薦方法 6
一、 基於受歡迎程度(Popularity-based)的推薦技術 : 7
二、 基於內容(Content-based)的標籤推薦技術 : 7
三、 協同過濾(Collaborative Filtering)的標籤推薦技術 : 8
四、 Adapted PageRank(APR) 和 FolkRank(FR) : 8
五、 因式分解機(FM) 和 成對相互作用張量分解(PITF) : 8
六、 時間標籤使用模式(GIRPTM) : 9
七、 Based-Level Learning(BLL) : 10
第三節、 研究缺口 10
第三章、 以序列模式為基礎之個人化標籤推薦(SP-PTR) 12
第一節、 標籤權重給定 13
一、 標籤候選集集成 : 13
二、 個人化標籤權重計算 : 13
三、 大眾化標籤權重計算 : 15
第二節、 最佳標籤選取 16
第四章、實證評估 17
第一節、 資料蒐集 17
第二節、 實驗設計 18
第三節、 評估方法 19
第四節、 評估指標 20
第五節、 實證分析與討論 22
一、 整體標籤推薦方法效能比較 : 23
二、 實驗結果之細部分析 – 只探討其他使用者的平均標籤權重 : 27
三、 實驗結果之細部分析 –只探討目標使用者的標籤權重 : 30
四、 實驗結果之細部分析 – 最佳推薦效能 : 35
第五章、結論與未來研究 40
參考文獻 42

Asanov Daniar, "Algorithms and methods in recommender systems," Berlin Institute of Technology, Berlin, Germany, 2011.
Basu Chumki, Hirsh Haym, and Cohen William, "Recommendation as classification: Using social and content-based information in recommendation," in Aaai/iaai, pp. 714-720, 1998
Batagelj Vladimir and Zaveršnik Matjaž, "Generalized cores," arXiv preprint cs/0202039, 2002.
Cattuto Ciro et al., "Network properties of folksonomies," Ai Communications, vol. 20, no. 4, pp. 245-262, 2007.
Cheng Tsang-Hsiang and Lee Yen-Hsien, "Improving Recommendation Performance with User Interest Evolution Patterns," in PACIS, p. 298, 2014
Ebbinghaus Hermann, "Memory: A contribution to experimental psychology," Annals of neurosciences, vol. 20, no. 4, p. 155, 2013.
García Ramón Rubio, Quirós Javier Suárez, Santos Ramón Gallego, González Santiago Martín, and Fernanz Samuel Morán, "Interactive multimedia animation with macromedia flash in descriptive geometry teaching," Computers & Education, vol. 49, no. 3, pp. 615-639, 2007.
Halpin Harry, Robu Valentin, and Shepherd Hana, "The complex dynamics of collaborative tagging," in Proceedings of the 16th international conference on World Wide Web, ACM, pp. 211-220, 2007
Hotho Andreas, Benz Dominik, Jäschke Robert, and Krause Beate, "ECML PKDD Discovery Challenge 2008 (RSDC’08)," in Workshop at 18th Europ. Conf. on Machine Learning (ECML’08)/11th Europ. Conf. on Principles and Practice of Knowledge Discovery in Databases (PKDD’08), vol. 32, 2008
Hotho Andreas, Jäschke Robert, Schmitz Christoph, and Stumme Gerd, "Information retrieval in folksonomies: Search and ranking," in European Semantic Web conference, Springer, pp. 411-426, 2006
Huang Sheng-Jun, Yu Yang, and Zhou Zhi-Hua, "Multi-label hypothesis reuse," in Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining, ACM, pp. 525-533, 2012
Huang Sheng-Jun and Zhou Zhi-Hua, "Fast multi-instance multi-label learning," arXiv preprint arXiv:1310.2049, 2013.
Huang Sheng-Jun, Zhou Zhi-Hua, and Zhou Z, "Multi-Label Learning by Exploiting Label Correlations Locally," in AAAI, pp. 949-955, 2012
Jäschke Robert, Marinho Leandro, Hotho Andreas, Schmidt-Thieme Lars, and Stumme Gerd, "Tag recommendations in folksonomies," in European Conference on Principles of Data Mining and Knowledge Discovery, Springer, pp. 506-514, 2007
Jäschke Robert, Marinho Leandro, Hotho Andreas, Schmidt-Thieme Lars, and Stumme Gerd, "Tag recommendations in social bookmarking systems," Ai Communications, vol. 21, no. 4, pp. 231-247, 2008.
Kroski Ellyssa, "The hive mind: Folksonomies and user-based tagging," InfoTangle Blog, December, vol. 9, 2005.
Lambiotte Renaud and Ausloos Marcel, "Collaborative tagging as a tripartite network," in International Conference on Computational Science, Springer, pp. 1114-1117, 2006
Li Junhuai, Wang Jinqin, Yu Lei, and Zhang Jing, "A novel frequent trajectory mining method based on GSP," in International Conference on Web Information Systems and Mining, Springer, pp. 134-140, 2011
Lipczak Marek, "Hybrid tag recommendation in collaborative tagging systems," 2012.
Marinho Leandro Balby et al., Recommender systems for social tagging systems. Springer Science & Business Media, 2012.
Marinho Leandro Balby et al., "Social tagging recommender systems," in Recommender systems handbook: Springer, pp. 615-644, 2011.
Marinho Leandro Balby and Schmidt-Thieme Lars, "Collaborative tag recommendations," in Data Analysis, Machine Learning and Applications: Springer, pp. 533-540, 2008.
Mathes Adam et al., "Folksonomies-Cooperative Classification and Communication Through Shared Metadata," D-Lib Magazine, vol. 6897, pp. 81-90, 2007.
Merholz Peter, "Ethnoclassification and vernacular vocabularies," Blog. Peterme. com. http://www. peterme. com/archives/000387. html, 2004.
Mishne Gilad, "Autotag: a collaborative approach to automated tag assignment for weblog posts," in Proceedings of the 15th international conference on World Wide Web, ACM, pp. 953-954, 2006
Musto Cataldo, Narducci Fedelucio, Lops Pasquale, and de Gemmis Marco, "Combining collaborative and content-based techniques for tag recommendation," in International Conference on Electronic Commerce and Web Technologies, Springer, pp. 13-23, 2010
Peterson Elaine, "Beneath the metadata: Some philosophical problems with folksonomy," D-Lib Magazine, vol. 12, no. 11, 2006.
Quintarelli Emanuele, "Folksonomies: power to the people," 2005.
Rattenbury Tye, Good Nathaniel, and Naaman Mor, "Towards automatic extraction of event and place semantics from flickr tags," in Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval, ACM, pp. 103-110, 2007
Read Jesse, Pfahringer Bernhard, Holmes Geoff, and Frank Eibe, "Classifier chains for multi-label classification," Machine learning, vol. 85, no. 3, p. 333, 2011.
Rendle Steffen, "Factorization machines," in Data Mining (ICDM), 2010 IEEE 10th International Conference on, IEEE, pp. 995-1000, 2010
Rendle Steffen and Schmidt-Thieme Lars, "Pairwise interaction tensor factorization for personalized tag recommendation," in Proceedings of the third ACM international conference on Web search and data mining, ACM, pp. 81-90, 2010
Shepitsen Andriy, Gemmell Jonathan, Mobasher Bamshad, and Burke Robin, "Personalized recommendation in social tagging systems using hierarchical clustering," in Proceedings of the 2008 ACM conference on Recommender systems, ACM, pp. 259-266, 2008
Sigurbjörnsson Börkur and Van Zwol Roelof, "Flickr tag recommendation based on collective knowledge," in Proceedings of the 17th international conference on World Wide Web, ACM, pp. 327-336, 2008
Sood Sanjay, Owsley Sara, Hammond Kristian J, and Birnbaum Larry, "TagAssist: Automatic Tag Suggestion for Blog Posts," in ICWSM, 2007
Symeonidis Panagiotis, Nanopoulos Alexandros, and Manolopoulos Yannis, "Tag recommendations based on tensor dimensionality reduction," in Proceedings of the 2008 ACM conference on Recommender systems, ACM, pp. 43-50, 2008
Trattner Christoph, Kowald Dominik, Seitlinger Paul, Ley Tobias, and Kopeinik Simone, "Modeling Activation Processes in Human Memory to Predict the Use of Tags in Social Bookmarking Systems," J. Web Science, vol. 2, no. 1, pp. 1-16, 2016.
Tso-Sutter Karen HL, Marinho Leandro Balby, and Schmidt-Thieme Lars, "Tag-aware recommender systems by fusion of collaborative filtering algorithms," in Proceedings of the 2008 ACM symposium on Applied computing, ACM, pp. 1995-1999, 2008
Van Maanen Leendert and Marewski Julian N, "Recommender systems for literature selection: A competition between decision making and memory models," in Proceedings of the 31st Annual Conference of the Cognitive Science Society, pp. 2914-2919, 2009
Vanderwal Thomas, "Off the top: Folksonomy entries," Visited November, vol. 5, no. 2005, p. 20, 2005.
Voss Jakob, "Tagging, folksonomy & co-renaissance of manual indexing?," arXiv preprint cs/0701072, 2007.
Weinberger D, "Controlled and suggested vocabularies: Are tags making us dumb," Joho the Blog [blog], 2005.
Zhang Lei, Tang Jian, and Zhang Ming, "Integrating temporal usage pattern into personalized tag prediction," in Asia-Pacific Web Conference, Springer, pp. 354-365, 2012
Zhang Min-Ling and Zhang Kun, "Multi-label learning by exploiting label dependency," in Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining, ACM, pp. 999-1008, 2010
Zhang Zi-Ke, Zhou Tao, and Zhang Yi-Cheng, "Tag-aware recommender systems: a state-of-the-art survey," Journal of computer science and technology, vol. 26, no. 5, p. 767, 2011.