結合意見探勘之電影推薦系統的研究

A Study of Movie Recommendation Systems with Opinion Mining

10.6846/TKU.2017.00565

陳仕堯

推薦系統 ； 意見探勘 ； 協同過濾 ； Movie Recommendation System ； Opinion Mining ； Collaborative-Filtering

淡江大學資訊管理學系碩士班學位論文

2017年

碩士

蕭瑞祥

繁體中文

由於在這資訊爆炸的時代中，網路上充斥著非常多的資訊，使用者需要瀏覽非常多的評論進行比較，最後再做出相關決策，因此許多學者開始研究相關資訊過濾的機制，而推薦系統就是屬於資訊過濾的一種，推薦系統可從大量的資訊中，可以依據使用者的年齡、性別或是需求，系統會記錄相關資訊並進行過濾，進而將資訊或是商品推薦給使用者，目前常見的推薦系統有內容過濾式推薦（Content-Based Filtering）、協同過濾式（Collaborative-Filtering）以及混合過濾式（Hybrid-Based Filtering）。 目前一般電影網站僅提供電影整體星等評價、預告片、導演等相關電影資訊，而民眾在觀賞電影前除了會參考整體的星等評價外，還會參考網路上他人的評論，才會決定是否觀賞電影，為了減少使用者在搜尋過程中付出額外時間成本，本研究提出一個結合意見探勘的電影推薦系統（以下簡稱結合意見探勘型），於系統背景中，蒐集使用者記錄，透過自動化、意見單元定義擷取等步驟收集IMDb網友的評論，對電影各個面向進行網路輿情分析，給予每部電影權重分數並進行推薦，最後將結合意見探勘型與傳統型電影推薦系統（以下簡稱一般型）兩者作比較，並探討使用者之使用意願與推薦準確率來評估本研究之成果。 本研究利用問卷調查、收集使用者紀錄進行推薦系統評估並得知近八成民眾滿意結合意見探勘型，且從推薦系統評估的結果中得知F-Measure為68.06%比起一般型高出7.54%，其推薦準確率達到82.84%，可以發現本研究提出的結合意見探勘型之推薦結果較符合使用者心中意願。

As the shift toward the big data continues, the internet is continued to be filled with amounts of information. Users are required to browse through these unfiltered data to make relevant decisions. So many researchers begin to study the classification of information filtering, such as a recommendation system. The system can record information and filter based on the age, gender, or needs of the user from a large amount of information . The more specific the criteria, the more accurate the results will be. There are currently three widely used recommendation systems, Content-Based Filtering, Collaborative Filtering, and the Hybrid-based. In the past, film reviewers only provided an overall score on the films reviewed. Now, consumers not only take film score into consideration, but also comments and opinions submitted by other users online. Therefore, this study proposes a collaborative filtering recommendation system,"A film recommendation system that incorporates consumer opinions(hereinafter referred to as the traditional method with opinion mining), and explores the details of the film recommended system. Then for each movie element analysis and give each film weight score and recommended. Finally, let traditional method with opinion mining and traditional film recommendation system (hereinafter referred to as the traditional method) for comparison to make users more in line with the user's expectations. From the results of our study, we found that by incorporating opinion mining, the F-Measure was 68.06%, which was 7.54% higher than that of the standard method. Compared with the standard film recommendation system, the system incorporating opinion mining is more in line with the opinions of the users.

1. [1]何文程（2010），奧斯卡得獎男女演員對電影票房收入效應之探討， 暨南大學經營管理碩士論文。
   連結：
2. [2]李孟潔（2009），利用機器學習作法之中文意見分析，清華大學資訊工程學系碩士論文。
   連結：
3. [3]李日斌（2014），探討臺灣網民對鄰國的情感，中山大學資訊管理學系碩士論文。
   連結：
4. [6]謝金育（2012），結合貝氏網路與激勵理論之推薦機制-電影推薦系統設計，國立交通大學資訊管理學系碩士論文。
   連結：
5. [8]陳聖昆（2014），電影元素對於消費者觀看意願之影響，淡江大學企業管理學系碩士論文。
   連結：
6. [9]黃心宜（2014），基於影響力分析之意見單元評價的研究，淡江大學資訊管理學系碩士論文。
   連結：
7. [11]楊盛帆（2009），以整合式規則來做網路論壇上的 3C 產品口碑分析，元智大學資訊管理學系碩士論文。
   連結：
8. [12]葉庭瑋（2013），基於意見詞修飾關係之微網誌情感分析技術，臺北科技大學資訊工程系研究所碩士論文，第二十五屆自然語言與語音處理研討會, 168-182。
   連結：
9. [14]Adomavicius, G., & 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, 17（6）, 734-749.
   連結：
10. [18]Bloom, K., & Argamon, S.（2009）. Automated learning of appraisal extraction patterns. Language and Computers, 71（1）, 249-260.
    連結：
11. [22]Dawes, J.,（2008）. “Do data characteristics change according to the number of scale points used? An experiment using 5-point, 7-point and 10-point scales ,” International Journal of Market Research. 50(1):61-77.
    連結：
12. [23]Eriksson, B.（2006）. Sentiment classification of movie reviews using linguistic parsing. Natural Language Processing. CS, 838.
    連結：
13. [24]Goldberg, D., Nichols, D., Oki, B. M., & Terry, D.（1992）. Using collaborative filtering to weave an information tapestry. Communications of the ACM, 35（12）, 61-70.
    連結：
14. [33]Nick, Z. Z., & Themis, P.（2001）. Web search using a genetic algorithm. IEEE Internet computing, 5（2）, 18.
    連結：
15. [35]Nunamaker Jr, J. F., Chen, M., & Purdin, T. D.（1990）. Systems development in information systems research. Journal of management information systems, 7（3）, 89-106.
    連結：
16. [47]Tottie, Gunnel.（1991） “Negation in English Speech and Writing: A Study in Variation,” San Diego: Academic Press.
    連結：
17. 一、中文文獻
18. [4]林銘嬋（2016），結合意見探勘之推薦系統，東吳大學資訊管理學系碩士論文。
19. [5]曹盛涵（2014），結合評論意見探勘之推薦系統，東吳大學資訊管理學系碩士論文。
20. [7]陳柏任（2012），運用社會網絡分析法以發掘興趣導向虛擬社群關鍵使用者之研究，國立高雄大學資訊管理學系碩士論文。
21. [10]黃彥瑋（2015），挖掘續集電影評論意見之研究，東吳大學資訊管理學系碩士論文。
22. 二、英文文獻
23. [13]Adomavicius, G., & Tuzhilin, A.（2004）. Recommendation Technologies: Survey of Current Methods and Possible Extensions (Working Paper). Stern School of Business, New York University.
24. [15]Balahur, A., Hermida, J. M., & Montoyo, A.（2011）. Detecting implicit expressions of sentiment in text based on commonsense knowledge. In Proceedings of the 2nd Workshop on Computational Approaches to Subjectivity and Sentiment Analysis（pp.53-60）.Association for Computational Linguistics.
25. [16]Barbosa, L., & Feng, J.（2010）. Robust sentiment detection on twitter from biased and noisy data. In Proceedings of the 23rd International Conference on Computational Linguistics: Posters（pp.36-44）. Association for Computational Linguistics.
26. [17]Basu, C., Hirsh, H., & Cohen, W.（1998）. Recommendation as classification: Using social and content-based information in recommendation. In Aaai/iaai （pp. 714-720）.
27. [19]Bobadilla, J., Ortega, F., Hernando, A., & Gutiérrez, A.（2013）. Recommender systems survey. Knowledge-based systems, 46, 109-132.
28. [20]Breese, J. S., Heckerman, D., & Kadie, C.（1998）. Empirical analysis of predictive algorithms for collaborative filtering. In Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence（pp. 43-52）. Morgan Kaufmann Publishers Inc..
29. [21]Chen, Y. H. & Wang, J. W.（2016）. Using Multiple Representations to Select Instances for Text Classification. The 2016 Conference on Computational Linguistics and Speech Processing ROCLING, 194-203.
30. [25]Hu, M., & Liu, B.（2004）. Mining and summarizing customer reviews. In Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining （pp. 168-177）. ACM.
31. [26]Ji, H., Li, J., Ren, C., & He, M.（2013）. Hybrid collaborative filtering model for improved recommendation. In Service Operations and Logistics, and Informatics (SOLI), 2013 IEEE International Conference on (pp. 142-145). IEEE.
32. [27]Kearney, P., Anand, S. S., & Shapcott, M. （2005）. Employing a domain ontology to gain insights into user behaviour. In Working Notes of the IJCAI Workshop on Intelligent Techniques for Web Personalization （pp. 25-32）.
33. [28]Kerlinger, F. N.,（1986）, Foundations of behavioral research, 3rd Edition, New York: McGraw-Hill
34. [29]Kim, S. M., & Hovy, E. (2006). Extracting opinions, opinion holders, and topics expressed in online news media text. In Proceedings of the Workshop on Sentiment and Subjectivity in Text （pp.1-8）.Association for Computational Linguistics.
35. [30]Likert, R.（1932）. “A Technique for the Measurement of Attitudes,” Archives of Psychology, Vol. 22, No. 140, pp. 1-55.
36. [31]Liu, B., Hu, M., & Cheng, J.（2005）. Opinion observer: analyzing and comparing opinions on the web. In Proceedings of the 14th international conference on World Wide Web (pp. 342-351). ACM.
37. [32]Miller, B. N., Albert, I., Lam, S. K., Konstan, J. A., & Riedl, J.（2003）. MovieLens unplugged: experiences with an occasionally connected recommender system. In Proceedings of the 8th international conference on Intelligent user interfaces （pp. 263-266）. ACM.
38. [34]Nunnally, J. C., & Bernstein, I. H.（1978）.Psychometric Theory. 2nd Edition, McGraw Hill, New York.
39. [36]O’connor, M., Cosley, D., Konstan, J. A., & Riedl, J. (2001). PolyLens: a recommender system for groups of users. In ECSCW 2001 (pp. 199-218). Springer Netherlands.
40. [37]Porter, M. F.（2001）.Snowball: A language for stemming algorithms, http://snowball.tartarus.org/texts/introduction.html.
41. [38]Resnick, P., Iacovou, N., Suchak, M., Bergstrom, P., & Riedl, J.（1994）. GroupLens: an open architecture for collaborative filtering of netnews. In Proceedings of the 1994 ACM conference on Computer supported cooperative work （pp. 175-186）. ACM.
42. [39]Santorini, B.（1990）. “Part-of-speech Tagging Guidelines for the Penn Treebank Project.” Tech. rep. MS-CIS-90-47, Department of Computer and Information Science, University of Pennsylvania.
43. [40]Sarwar, B., Karypis, G., Konstan, J., & Riedl, J.（2000a）. Analysis of recommendation algorithms for e-commerce. In Proceedings of the 2nd ACM conference on Electronic commerce（pp. 158-167）. ACM.
44. [41]Sarwar, B., Karypis, G., Konstan, J., & Riedl, J.（2000b）. Application of dimensionality reduction in recommender system-a case study（No. TR-00-043）. Minnesota Univ Minneapolis Dept of Computer Science.
45. [42]Sarwar, B., Karypis, G., Konstan, J., & Riedl, J.（2001）. Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th international conference on World Wide Web（pp. 285-295）. ACM.
46. [43]Schafer, J. B., Konstan, J., & Riedl, J.（1999）. Recommender systems in e-commerce. In Proceedings of the 1st ACM conference on Electronic commerce（pp. 158-166）. ACM.
47. [44]Schafer, J. B., Konstan, J. A., & Riedl, J.（2001）. E-commerce recommendation applications Applications of Data Mining to Electronic Commerce （pp. 115-153）: Springer.
48. [45]Sokolova, M., Japkowicz, N., & Szpakowicz, S.（2006）. Beyond accuracy, F-score and ROC: a family of discriminant measures for performance evaluation. In Australasian Joint Conference on Artificial Intelligence （pp. 1015-1021）. Springer Berlin Heidelberg.
49. [46]Wiebe, J., & Riloff, E.（2005）. Creating subjective and objective sentence classifiers from unannotated texts. In International Conference on Intelligent Text Processing and Computational Linguistics（pp. 486-497）. Springer Berlin Heidelberg.
50. [48]Zhuang, L., Jing, F., & Zhu, X. Y.（2006）. Movie review mining and summarization. In Proceedings of the 15th ACM international conference on Information and knowledge management (pp. 43-50). ACM.