臺灣博碩士論文加值系統

隨著網路技術的發達，如果想要找尋資料，第一個想到的一定是透過網路，透過網路搜尋引擎，可以搜尋到大量的資訊，但這些過多的資訊卻造成了「資訊過載」的情況，因此，在本研究的系統中，建立了一個「精華文章」的區塊，去除了大量的廣告、網路書店的資訊，只留下有關於教學的文章，來提供使用者閱讀。
本研究提出一套混合式推薦的方法來給予學習者適性化的推薦，混合式推薦是利用「內容式推薦」與「協同過濾推薦」這兩種方式來做推薦。本研究先針對「內容式推薦」的部分，先利用Term Frequency - Inverse Document Frequency來計算文章的特徵值，再經由Neural Text Categorizer來訓練，在訓練的過程中，利用本研究提出的詞庫修正方法，來擴充與修正詞庫的權重，在訓練完成後，利用詞庫權重加總來作為推薦的分數。在「協同式過濾」的部分，利用User-based的方式，取出相似度較高的使用者喜好來做計算，再與本研究提出群組喜好推薦的方式結合，針對群組(Group)中的特徵來做分析，調整每位使用者的推薦項目，以符合使用者的喜好。
本研究透過混合式推薦，提供給學習者適性化的推薦，由網路文章提取出教學文章的效益皆在90%以上，並在公開資料集上驗證喜好推薦的方式，能提高10%的命中率，再應用到平台中，依每位學習者不同的能力、興趣給予推薦的文章，讓每位學習者能夠閱讀符合需求的文章。

Along with advanced network technology, if you want to search for information, the first thought must be through the network. Use the search engine, you can find a lot of information, but it caused “information overload”. Therefore, in the study system created "essence of the article" block, automatically removes a lot of advertising and online bookstore information, leaving only about teaching the article.
A hybrid recommended method to give learners adaptive recommendation is proposed in this study; hybrid recommended is to use ‘Content-based recommendation’ and ‘Collaborative filtering’ these two methods to recommend. The Content-based in this study, Term Frequency - Inverse Document Frequency (TF-IDF) method is used to calculate the feature of articles. And then, use the Neural Text Categorizer (NTC) to training keyword bank. In the training process, this study proposes adjustment method to extend and revise the weights in keyword bank. After training is completed, use this keyword bank weights as recommended score. The Collaborative filtering in this study, use User-based to do calculations user preferences with high similarity users, and then combine ‘Group recommendation’ to analysis group features and adjustment user’s recommend items with user’s preferences.
This research used hybrid recommendation to provide learners adaptive recommendation, according to each learner has different abilities and interests to give the recommended articles, so that each learner can read article meet their needs. The experimental results show a high accuracy in the recommendations articles.

摘　　要 iv
ABSTRACT v
致　　謝 vii
目　　次 viii
表目錄 x
圖目錄 xi
第一章 緒論 1
1.1 研究動機 1
1.2 研究目的 2
1.3 論文架構 3
第二章 文獻探討 4
2.1 本文分類 4
2.1.1 N-Gram斷詞 4
2.1.2 Term Frequency - Inverse Document Frequency 6
2.1.3 Neural Text Categorizer 7
2.2 推薦系統 9
2.2.1 內容式推薦系統 9
2.2.2 協同過濾推薦系統 10
2.2.3 混合式推薦系統 13
第三章 研究方法 15
3.1 系統架構 15
3.2 內容式文章推薦方法 17
3.2.1. 文章前處理 17
3.2.2. 初始詞庫 18
3.2.3. 訓練模組 20
3.2.4. 詞庫擴充修正 24
3.2.5. 分類模組 26
3.2.6. 推薦模組 28
3.3 協同過濾式推薦方法 29
3.3.1. User-based Collaborative Filtering 29
3.3.2. Item-Based Collaborative Filtering 30
3.3.3. 群組喜好推薦方法 31
3.4 混合式推薦方法 32
3.4.1. 推薦方法 32
3.4.2. 使用者回饋 33
第四章 實驗與結果討論 34
4.1 實驗資料 34
4.1.1 內容式推薦資料集 34
4.1.2 協同式過濾推薦資料集 36
4.2 評估準則 37
4.3 實驗結果與分析 40
4.3.1 實驗一：詞庫正負向比例與分類效果評估 40
4.3.2 實驗二：群組喜好推薦方法於公開資料集上驗證 47
第五章 結論與未來展望 51
參考文獻 53
附錄 57
A. 電影類別與影片數 57
B. 類別間相似度排序 58

1.Jerome R. B. (2000), Exploiting Latent Semantic Information in Statistical Language Modeling, Proceedings of the IEEE, Vol. 88, No. 8, pp. 1279-1296.
2.Qiang Y. and Zhang, H.H. (2003), Web-Log Mining for Predictive Web Caching, IEEE Transactions on Knowledge and Data Engineering, Vol. 15, No. 4, pp. 1050-1053.
3.Suleiman H. M. (2005), Character contiguity in N-gram-based word matching: the case for Arabic text searching, Information Processing and Management, Vol. 41, No. 4, pp. 819-827.
4.Rani, B.P. , Vardhan, B.V. , Durga, A.K. , Reddy, L.P. and Babu, A.V. (2008), Analysis of N-gram model on Telugu Document Classifiaction, IEEE Congress on Evolutionary Computation(CEC 2008),pp. 3199-3203.
5.Mike C., Son D., Ai K. and Nigel C. (2009), Classifying disease outbreak reports using n-grams and semantic features, International Journal of Medical Informatics, Vol. 78, No. 12, pp. e47-e58.
6.Makoto S., Naohide Y., Yi-Ching T., Takashi Ishida and Masayuki G. (2010), English And Taiwanese Text Categorization Using N-gram Based on Vector Space Model, ISITA2010, Taichung, Taiwan, October 17-20.
7.Naptali W., Tsuchiya M. and Nakagawa S. (2012), Topic-Dependent-Class-Based n-gram Language Model, IEEE Transactions on Audio, Speech, and Language Processing, Vol. 20, No. 5, pp.1513-1525.
8.Alzahrani S.M., Salim N. and Abraham A. (2012) Understanding Plagiarism Linguistic Patterns, Textual Features, and Detection Methods, IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Vol. 42, No. 2, pp.133-149.
9.O'Kane P., Sezer S. , McLaughlin K. and Eul Gyu Im (2013), SVM Training Phase Reduction Using Dataset Feature Filtering for Malware Detection, IEEE Transactions on Information Forensics and Security, Vol. 8, No. 3, pp.500-509.
10.Wu HC, Luk RWP, Wong KF, Kwok KL (2008), Interpreting tf–idf term weights as making relevance decisions, ACM Transactions on Information Systems, Vol. 26, No. 3,pp.13:1-13:36.
11.Salton G, Buckley C (1988), Term-weighting approaches in automatic text retrieval, Information Processing & Management, Vol. 24, No. 5, 1988, pp. 513-523.
12.Taeho Jo (2008), Neural Text Categorizer for Exclusive Text Categorization, Journal of Information Processing Systems, Vol. 4, No. 2, pp.77-86.
13.Linden G., Smith B. and York J. (2003), Amazon.com Recommendations Item-to-Item Collaborative Filtering, IEEE Internet Computing, Vol. 7, No. 1,pp.76-80.
14.Rokach L. and Kisilevich S. (2012), Initial Profile Generation in Recommender Systems Using Pairwise Comparison, IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Vol. 42, No. 6, pp.1854-1859.
15.Keunho Choi and Yongmoo Suh (2013), A new similarity function for selecting neighbors for each target item in collaborative filtering, Knowledge-Based Systems, Vol. 37, pp. 146-153.
16.Edward Rolando Núñez-Valdéz, Juan Manuel Cueva Lovelle, Oscar Sanjuán Martínez, Vicente García-Díaz, Patricia Ordoñez de Pablos and Carlos Enrique Montenegro Marín (2012), Implicit feedback techniques on recommender systems applied to electronic books, Computers in Human Behavior , Vol. 28, No. 4, pp.1186-1193.
17.Zhaojun Yang, Levow G.A. and Meng H. (2012), Predicting User Satisfaction in Spoken Dialog System Evaluation With Collaborative Filtering, IEEE Journal of Selected Topics in Signal Processing, Vol. 6, No. 8, pp.971-981.
18.Inma Garcia, Sergio Pajares, Laura Sebastia and Eva Onaindia (2012), Preference elicitation techniques for group recommender systems, Information Sciences, Vol. 189, No. 15, pp.155-175.
19.Pei-Yu Wang and Hui-Chun Yang (2012), Using collaborative filtering to support college students’ use of online forum for English learning, Computers & Education, Vol. 59, No. 2, pp.628-637.
20.Qi Liu, Enhong Chen , Hui Xiong , Ding C.H.Q. and Jian Chen (2012), Enhancing Collaborative Filtering by User Interest Expansion via Personalized Ranking, IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, Vol. 42, No. 1, pp.218-233.
21.Robin Burke (2002), Hybrid Recommender Systems Survey and Experiments., User Modeling and User-Adapted Interaction, Vol. 12, No. 4, pp.331-370.
22.Adomavicius G. and Tuzhilin A. (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, Vol. 17, No. 6, pp.734-749.
23.Yoshii K., Goto M., Komatani K. , Ogata, T. and Okuno H.G. (2008), An Efficient Hybrid Music Recommender System Using an Incrementally Trainable Probabilistic Generative Model, IEEE Transactions on Audio, Speech, and Language Processing, Vol. 6, No. 2, pp.435-447.
24.Martinez A., Pazos Arias J.J., Vilas A.F., Duque J.G. and Nores M.L. (2009), What's on TV Tonight? An Efficient and Effective Personalized Recommender System of TV Programs, IEEE Transactions on Consumer Electronics, Vol. 55, No. 1, pp.286-294.
25.Subhash K. Shinde and Uday Kulkarni (2012), Hybrid personalized recommender system using centering-bunching based clustering algorithm, Expert Systems with Applications, Vol. 39, No. 1, pp.1381-1387.
26.Jürgen Buder and Christina Schwind (2012), Learning with personalized recommender systems: A psychological view, Computers in Human Behavior, Vol. 28, No. 1, pp.207-216.
27.Walter Carrer-Neto, María Luisa Hernández-Alcaraz, Rafael Valencia-García , and Francisco García-Sánchez (2012), Social knowledge-based recommender system. Application to the movies domain, Expert Systems with Applications, Vol. 39, No. 12, pp.10990-11000.
28.Ahmad A. Kardan and Mahnaz Ebrahimi (2013), A novel approach to hybrid recommendation systems based on association rules mining for content recommendation in asynchronous discussion groups, Information Sciences, Vol. 219, pp.93-110.
29.Adomavicius G. and YoungOk Kwon (2012), Improving Aggregate Recommendation Diversity Using Ranking-Based Techniques, IEEE Transactions on Knowledge and Data Engineering, Vol. 24, No. 5, pp.896-911.
30.Wikipedia,” Stop words”, https://en.wikipedia.org/wiki/Stop_words.
31.陳重光，應用雲端虛擬化技術結合資訊檢索之演算法教學文章推薦，南台科技大學資訊工程系碩士論文，民101年。
32.陳立人，利用Latent Dirichlet Allocation之個人化文章推薦，國立中山大學資訊管理學系碩士論文，民101年。
33.吳振銘，應用改良式K-means分群法於個人化音樂推薦服務系統之實現，國立高雄應用科大電子工程系碩士論文，民101年。
34.賴昆佑，以統計分析探討文件分類程序對期刊論文分類效果之影響，國立中央大學資訊管理研究所碩士論文，民96年。
35.蔡忠霖，應用於多源基因體學分類問題之以熵值為基礎的特徵選取法，國立成功大學資訊管理研究所碩士論文，民96年。
36.林重佑，學習向量量化式交換策略混合過濾電影推薦架構，國立雲林科技大學資訊管理碩士班碩士論文，民101年。