進階搜尋


   電子論文尚未授權公開,紙本請查館藏目錄
(※如查詢不到或館藏狀況顯示「閉架不公開」,表示該本論文不在書庫,無法取用。)
系統識別號 U0026-2508201620413800
論文名稱(中文) 結合主題地圖與社群影響之個人化混合式推薦
論文名稱(英文) Adapting Topic Map and Social Influence to the Personalized Hybrid Recommender System
校院名稱 成功大學
系所名稱(中) 資訊管理研究所
系所名稱(英) Institute of Information Management
學年度 104
學期 2
出版年 105
研究生(中文) 蔡育姍
研究生(英文) Yu-Shan Tsai
學號 r76034064
學位類別 碩士
語文別 英文
論文頁數 53頁
口試委員 指導教授-王惠嘉
口試委員-高宏宇
口試委員-劉任修
口試委員-盧文祥
中文關鍵字 混合式推薦系統  冷起始  社群網路  知識本體論  情感分析 
英文關鍵字 Hybrid Recommender System  Cold Start  Social Network  Ontology  Sentiment Analysis 
學科別分類
中文摘要 由於Web 2.0的成熟發展,使用者習慣透過電子口碑去判斷商家的好壞。然而大量的使用者生成內容導致使用者難以從中發掘出所需資訊。推薦系統作為一個資訊擷取工具,能夠有效地幫助使用者快速取得相關資訊,但目前推薦系統大多依靠項目的整體評分或熱門程度去作推薦,並沒有考慮到使用者為項目所表達的情感資訊,使得推薦結果難以符合使用者個人喜好。

另一方面,推薦系統多半面臨兩個主要的問題:冷起始和模型可擴展性。冷起始即系統的起始評分資訊不足以預測使用者偏好;模型可擴展性則指系統是否能因應資料維度過高的情況。推薦系統的表現容易受到兩者的影響,使得推薦結果不符合使用者的需求‧

因此本論文希望能夠建立一套推薦系統,一方面藉由使用者的社群關係,包含利用情感分析所得的喜好相似關係、朋友關係和信任關係等,協助預測使用者對於商家的評價,同時降低冷起始對於推薦表現的負面影響;另一方面,本研究利用知識本體論將取得的特徵整合成具代表性的主題,藉此對評論資料進行降維,以提高整體運算速度,並將每間商家被提及的主題以主題地圖的方式呈現,讓使用者可以參考相似商家的評價。

此外,本論文亦針對所提出的推薦演算法設計數個實驗,以驗證方法之有效性。在推薦表現上,可以觀察到藉由社群影響力去預測使用者的可能評分,多數分類表現大於.8,且分數預測差異值約略落於1分附近,表示利用使用社群關係有助於有效預測使用者對於項目的偏好,其中以社群關係的組合表現較佳。
英文摘要 The excessive amount of user-generated reviews results in the difficulty of extracting the relevant information. A recommender system is a solution to help users get the accurate information efficiently. Most of the existing recommender systems neglected emotions expressed by users in the reviews. It caused the failure of predicting user preferences accurately. On the other hand, the cold start problem and the model scalability are the two thorny problems to the recommender system. Cold start exists while lacking initial ratings; and model scalability is the capability of a model to cope with the high-dimensional data. These problems may mislead the recommendation, and users are not satisfied with the results accordingly.

A personalized recommender system is proposed to mitigate the negative effects the aforementioned problems cause. After extracting user preference, the social influence network is built accordingly. The predicted ratings are estimated based on the importance of users. Also, ontologies are applied to integrate the extracted features into topics for the sake of dimensionality reduction; and topics mentioned in the reviews are displayed as a form of topic map.

In addition, this thesis designed a number of experiments to validate the effectiveness of the proposed method. RMSEs and MAEs of all relationships are close 1, and most of the F1 measure is larger than .8. Both of them indicate that the proposed method is able to estimate the unknown ratings well with the help of social influence. Among all, the combinations of relationships perform better.
論文目次 Oral presentation document i
摘要ii
Abstract iii
Acknowledgments iv
List of Tables vii
List of Figures viii
Chapter 1. Introduction 1
1.1 Background and Motivations 1
1.2 Objectives 3
1.3 Scope and Limitations 4
1.4 Research Process 4
1.5 Research Overview 5
Chapter 2. Literature Review 7
2.1 Personalized Recommender System 7
2.1.1 Information Filtering 7
2.1.2 Cold Start Problem 8
2.2 Social Network Analysis 9
2.3 Natural Language Processing 10
2.3.1 Part-of-Speech Tagging 10
2.3.2 Vector Space Model 11
2.4 Sentiment Analysis 13
2.5 Topic Map 14
2.6 Ontology 15
2.7 Summary of the Related Works 15
Chapter 3. Research Methods 17
3.1 Theoretical Framework 17
3.2 Data Preprocessing Module 18
3.3 Topic Mapping Module 21
3.4 Sentiment Analysis Module 23
3.5 Hybrid Recommender Module 27
3.5.1 Nearest Neighbors Based on the Topic Similarity 27
3.5.2 Social Influence between Users 29
3.5.3 Item Recommendation based on Predicted Ratings 30
3.6 Summary of the Proposed Method 32
Chapter 4. Experiments and Analyses 33
4.1 Experimental Environment 33
4.2 Dataset 34
4.3 Experimental Setup 36
4.3.1 K-fold Cross Validation 36
4.3.2 Evaluation Metrics 36
4.3.3 Sample Size Determination 37
4.4 Parameter Settings 38
4.4.1 Assembling the Consecutive Nouns 38
4.4.2 Distance between a Feature and its Descriptive Terms 39
4.4.3 Integration of Features into Topics 40
4.4.4 Discovery of Neighbors with Similar Interest in Topics 41
4.5 Performance Comparison 42
Chapter 5. Conclusions 45
5.1 Contribution and Discussion 45
5.2 Future Research Work 46
References 48
參考文獻 Baccianella, S., Esuli, A., & Sebastiani, F. (2010). Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining. In 7th language resources and evaluation conference (lrec 2010) (pp. 2200–2204).
Bobadilla, J., Ortega, F., Hernando, A., & Bernal, J. (2012). A collaborative filtering approach to mitigate the new user cold start problem. Knowledge-Based Systems, 26, 225 - 238. Retrieved from http://www.sciencedirect.com/science/article/pii/S0950705111001882
BrightLocal. (2015). Local consumer review survey 2015 [Web Page]. BrightLocal. Retrieved from https://www.brightlocal.com/learn/local-consumer-review-survey/
Cambria, E., Schuller, B., Xia, Y., & Havasi, C. (2013). New avenues in opinion mining and sentiment analysis. Intelligent Systems, IEEE, 28(2), 15-21.
Cantallops, A. S., & Salvi, F. (2014). New consumer behavior: A review of research on ewom and hotels. International Journal of Hospitality Management, 36, 41 - 51. Retrieved from http://www.sciencedirect.com/science/article/pii/S0278431913001126
Choi, K., Yoo, D., Kim, G., & Suh, Y. (2012). A hybrid online-product recommendation system: Combining implicit rating-based collaborative filtering and sequential pattern analysis. Electronic Commerce Research and Applications, 11(4), 309 - 317. Retrieved from http://www.sciencedirect.com/science/article/pii/S156742231200018X
Cochran, W. G. (1977). Sampling techniques (3rd ed.). John Wiley & Sons. Dragut, E., & Fellbaum, C. (2014). The role of adverbs in sentiment analysis. In Proceedings of frame semantics in nlp: A workshop in honor of chuck fillmore (1929-2014) (pp. 38–41). Baltimore, MD, USA: Association for Computational Linguistics. Retrieved from http://www.aclweb.org/anthology/W14-3010
Escalante, H. J., García-Limón, M. A., Morales-Reyes, A., Graff, M., y Gómez, M. M., Morales, E. F., & Martínez-Carranza, J. (2015). Term-weighting learning via genetic programming for text classification. Knowledge-Based Systems, 83, 176 - 189. Retrieved from http://www.sciencedirect.com/science/article/pii/S0950705115001197
Fattah, M. A. (2015). New term weighting schemes with combination of multiple classifiers for sentiment analysis. Neurocomputing, 167, 434 - 442. Retrieved from http://www.sciencedirect.com/science/article/pii/S0925231215005202
Forsati, R., & Shamsfard, M. (2014). Hybrid pos-tagging: A cooperation of evolutionary and statistical approaches. Applied Mathematical Modelling, 38(13), 3193 - 3211. Retrieved from http://www.sciencedirect.com/science/article/pii/S0307904X1300783X
Golbeck, J. (2013). Chapter 6 - trust. In J. Golbeck (Ed.), Analyzing the social web (p. 75 - 89). Boston: Morgan Kaufmann. Retrieved from http://www.sciencedirect.com/science/article/pii/B9780124055315000067
Guo, G., Zhang, J., & Thalmann, D. (2014). Merging trust in collaborative filtering to alleviate data sparsity and cold start. Knowledge-Based Systems, 57, 57- 68. Retrieved from http://www.sciencedirect.com/science/article/pii/S0950705113003870
Hu, N., Koh, N. S., & Reddy, S. K. (2014). Ratings lead you to the product, reviews help you clinch it? the mediating role of online review sentiments on product sales. Decision Support Systems, 57, 42 - 53. Retrieved from http://www.sciencedirect.com/science/article/pii/S0167923613001942
Jeong, E., & Jang, S. S. (2011). Restaurant experiences triggering positive electronic word-of-mouth (ewom) motivations. International Journal of Hospitality Management, 30(2), 356 - 366. Retrieved from http://www.sciencedirect.com/science/article/pii/S0278431910000988
Kara, S., Alan, Ö., Sabuncu, O., Akpınar, S., Cicekli, N. K., & Alpaslan, F. N. (2012). An ontology-based retrieval system using semantic indexing. Information Systems, 37(4), 294 - 305. Retrieved from http://www.sciencedirect.com/science/article/pii/S030643791100113X (Semantic Web Data Management)
Linden, G., Smith, B., & York, J. (2003). Amazon.com recommendations: Itemto-item collaborative filtering. IEEE Internet Computing, 7(1), 76–80.
Lis, B. (2013). In ewom we trust. Business & Information Systems Engineering, 5(3), 129-140.
Liu, H., He, J., Wang, T., Song, W., & Du, X. (2013). Combining user preferences and user opinions for accurate recommendation. Electronic Commerce Research and Applications, 12(1), 14 - 23. Retrieved from http://www.sciencedirect.com/science/article/pii/S1567422312000312
Lu, B., Ott, M., Cardie, C., & Tsou, B. K. (2011). Multi-aspect sentiment analysis with topic models. In Proceedings of the 2011 ieee 11th international conference on data mining workshops (pp. 81–88). Washington, DC, USA: IEEE Computer Society.
Marcus, M. P., Marcinkiewicz, M. A., & Santorini, B. (1993). Building a large annotated corpus of english: The penn treebank. Comput. Linguist., 19(2), 313–330. Retrieved from http://dl.acm.org/citation.cfm?id=972470.972475
Navigli, R., & Ponzetto, S. P. (2012). BabelNet: The automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artificial Intelligence, 193, 217-250.
Nikolaev, A. G., Razib, R., & Kucheriya, A. (2015). On efficient use of entropy centrality for social network analysis and community detection. Social Networks, 40, 154 - 162. Retrieved from http://www.sciencedirect.com/science/article/pii/S0378873314000550
Özyer, T., Rokne, J., Wagner, G., & Reuser, A. H. (Eds.). (2013). The influence of technology on social network analysis and mining. Springer-Verlag Wien.
Peng, F., Lu, J., Wang, Y., Xu, R. Y. D., Ma, C., & Yang, J. (2016). N-dimensional markov random field prior for cold-start recommendation. Neurocomputing, 191, 187 - 199. Retrieved from http://www.sciencedirect.com/science/article/pii/S0925231216000825
Princeton University. (2010). About wordnet. Retrieved from http://wordnet.princeton.edu
Qian, X., Feng, H., Zhao, G., & Mei, T. (2014). Personalized recommendation combining user interest and social circle. Knowledge and Data Engineering, IEEE Transactions on, 26(7), 1763-1777.
Qiu, L., & Yuan, J. (2014). Construction of semantic associative network based on topic maps. In Broadband and wireless computing, communication and applications (bwcca), 2014 ninth international conference on (pp. 325–328).
Quan, C., & Ren, F. (2014). Unsupervised product feature extraction for feature-oriented opinion determination. Information Sciences, 272, 16 - 28. Retrieved from http://www.sciencedirect.com/science/article/pii/S0020025514001698
Shafiq, O., Alhajj, R., & Rokne, J. G. (2015). On personalizing web search using social network analysis. Information Sciences, 314, 55 - 76. Retrieved from http://www.sciencedirect.com/science/article/pii/S0020025515001292
Sinha, P., Veyie, N. M., & Purkayastha, B. S. (2015). Enhancing the performance of part of speech tagging of nepali language through hybrid approach. International Journal of Emerging Technology and Advanced Engineering, 5(5), 354 - 359. Retrieved from http://www.ijetae.com/files/Volume5Issue5/IJETAE_0515_61
Son, L. H. (2016). Dealing with the new user cold-start problem in recommender systems: A comparative review. Information Systems, 58, 87 - 104. Retrieved from http://www.sciencedirect.com/science/article/pii/S0306437914001525
Sun, J., Wang, G., Cheng, X., & Fu, Y. (2015). Mining affective text to improve social media item recommendation. Information Processing & Management, 51(4), 444 - 457. Retrieved from http://www.sciencedirect.com/science/article/pii/S0306457314000879
Thuan, T. T., & Puntheeranurak, S. (2014). Hybrid recommender system with review helpfulness features. In Tencon 2014 - 2014 ieee region 10 conference (pp. 1–5).
Toutanova, K., Klein, D., Manning, C. D., & Singer, Y. (2003). Feature-rich part-of-speech tagging with a cyclic dependency network. In Proceedings of the 2003 conference of the north american chapter of the association for computational linguistics on human language technology - volume 1 (pp. 173–180). Stroudsburg, PA, USA: Association for Computational Linguistics.
Yelp. (2014). Nielsen survey says …consumers rank yelp most influential, most trustworthy, and with highest quality reviews [Official web blog message]. Retrieved from https://www.yelpblog.com/2014/07/nielsen-survey-saysconsumers-rank-yelp-most-influential-most-trustworthy-and-with-highest-quality-re
Yelp. (2015). Yelp dataset challenge [Dataset]. Retrieved from http://www.yelp.com/dataset_challenge
Yuan, T., Cheng, J., Zhang, X., Liu, Q., & Lu, H. (2015). How friends affect user behaviors? an exploration of social relation analysis for recommendation. Knowledge-Based Systems, 88, 70 - 84. Retrieved from http://www.sciencedirect.com/science/article/pii/S0950705115003093
Zhang, J., Peng, Q., Sun, S., & Liu, C. (2014). Collaborative filtering recommendation algorithm based on user preference derived from item domain features. Physica A: Statistical Mechanics and its Applications, 396, 66 - 76. Retrieved from http://www.sciencedirect.com/science/article/pii/S0378437113010558
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2021-08-25起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw