基於貝氏網路之圖書推薦系統

A Book Recommender System Based on the Bayesian Network

10.6846/TKU.2016.00862

劉羣冠

貝氏網路 ； 協同推薦 ； 內容推薦 ； 推薦系統 ； Bayesian network ； collaborative recommendation ； content-based recommendation ； recommender system

淡江大學資訊管理學系碩士班學位論文

2016年

碩士

魏世杰

繁體中文

本文將實做一個推薦系統，透過貝氏網路結合協同與內容推薦方法，可以在沒有使用者背景的情況下，依據「給定的一本書」，進行相關圖書的推薦，希望能增進使用者查詢借閱書籍之效率與提高圖書館的書本借閱率。因為目前一般圖書館少有推薦系統，且有推薦系統的圖書館大多針對使用者進行推薦，這樣使用者必須登入系統後才可有推薦清單。但使用者通常習慣在查詢書本時不想登入，以省去麻煩及遭追蹤的可能，所以希望可以透過大眾化書推薦書的方式，讓系統在不取得使用者背景的情況下進行推薦。 本文透過離線實驗結果得出在貝氏網路架構下加入關鍵字與分類號節點資訊皆可提高其推薦結果且優於傳統推薦方法。本文也讓使用者進行線上實驗佐證本推薦方法的實用性。

In this work, we will present a recommender system which combines the collaborative and content-based methods using the Bayesian network. Given a book, it can provide related book recommendation without knowing the user identity so that users can look up the desired books more easily and the library books can be utilized more efficiently. Currently, few libraries provide a book recommender system. For those libraries which provide the recommender system, the user must login to receive the recommended book list. But users are accustomed to looking up the books without login to save the trouble or the risk of being tracked. This work aims to provide an anonymous way of book recommendation about a book without knowledge of the user identity. In our offline experiments, the results show that under the framework of the Bayesian network, adding the keyword nodes and the subject code nodes can help promote the recommendation performance when compared with the traditional recommendation methods. This work also conducts an online user experiment to demonstrate the usefulness of our recommendation method.

1. [3] 陳垂呈、陳幸暉，建置圖書館書籍推薦系統: 資料探勘之應用，工程科技與教育學刊 ，8(3)，469-478，2011。
   連結：
2. [4] 陳慶宇，「圖書館借書推薦系統之建置-以淡江圖書館資料為例」，淡江大學資訊管理學系碩士班學位論文，2014。
   連結：
3. [5] Balabanovic, M. & Shoham, Y., Fab: content-based, collaborative recommendation, Communications of the ACM, 40(3), 66-72, 1997.
   連結：
4. [8] Chan, L. M., Immroth's Guide to the Library of Congress Classification, (p. 19). Englewood, Col: Libraries Unlimited, 1990.
   連結：
5. [9] De Campos, L. M., Fernández-Luna, J. M., Huete, J. F. & Rueda-Morales, M. A., Combining conten-based and collaborative recommendations: a hybrid approach based on Bayesian networks, International Journal of Approximate Reasoning, 51 (7), 785-799, 2010.
   連結：
6. [11] Jaccard, J. & Turrisi, R., Interaction Effects in Multiple Regression, Sage, 2003.
   連結：
7. [13] Lin, L.I-K., A concordance correlation coefficient to evaluate reproducibility, Biometrics, 45 (1), 255-268, 1989.
   連結：
8. [14] Manning, C. D., Raghavan, P., Schütze, H, Introduction to information retrieval, Cambridge: Cambridge University Press, 2008.
   連結：
9. [16] Pearl, J. & Russell, S., Bayesian networks, Computer Science Department, University of California, 1998.
   連結：
10. [17] Ricci, F., Rokach, L., Shapira, B., Recommender Systems Handbook, Springer, 2015.
    連結：
11. [18] Sharif, M. A. & Raghavan, V. V., A large-scale, hybrid approach for recommending pages based on previous user click pattern and content, International Symposium on Methodologies for Intelligent Systems, 103-112, 2014.
    連結：
12. [1] 戴玉旻，「圖書館借閱紀錄探勘系統」，國立交通大學資訊科學研究所未發表碩士論文，2002。
13. [2] 賴永祥，中國圖書分類法，2007。
14. [6] Billsus, D. & Pazzani, M. J., Learning collaborative information filters, International Conference on Machine Learning, 98, 46-54, 1998.
15. [7] Burke, R. Knowledge-based recommender systems, Encyclopedia of Library and Information Science, 69 (32), 180–200, 2000.
16. [10] Gorakala, S. K. & Usuelli, M., Building a Recommendation System with R. Packt Publishing Ltd, 2015.
17. [12] Jannach, D., Zanker, M., Felfernig, A., Friedrich, D., Recommender Systems: an Introduction, Cambridge University Press, 2010.
18. [15] Owen, S., Anil, R., Dunning, T., Friedman, E, Mahout in action, Manning ,2012
19. [19] Tsai, C. H., MMSEG: A word identification system for Mandarin Chinese text based on two variants of the maximum matching algorithm, 2000, Available at http://www. geocities. com/hao510/mmseg.