臺灣博碩士論文加值系統

隨著網路購物的普及，消費者在買產品前逐漸提高對評論 依賴度。當越來容易被張貼於網路，產生捏造的評論欺騙消費者可能性。多研究開始關心虛假評論，但都僅限於單一領域而唯有跨辨別的究，在辨別跨領域評論效果不如預期。
故本研究設計出可檢測跨領域辨別真實與虛假評論的通用 性方法，使Stimuli-Organism-Response (S-O-R) 框架作為虛假評論 字詞的理構，並以此起點推架作為虛假評論 字詞的理構，並以此起點推有判斷力的字詞類別。藉由此發展出可以跨領域辨真實與虛假評論法則。本研究所發展之判斷法結果顯示，可以有效的辨別真實與虛假評論並在跨領域應用達到八成的精準度(Accuracy)。

With the rapid development of online shopping, consumers increasingly rely on user-generated online reviews to facilitate decision makings. As a result, deceptive review comments have been posted to elude readers. Most research concerns of this issues built up rules based on text data banks closely related to specific domains. To the best of our knowledge, only one study has strived to construct decision models to detect deceiving posts. However, the accuracy is only a bit higher than 50% when being applied to detect cross domain reviews.
This study devising a general method to detect deceptive review comments, by adapting S-O-R framework. On the basis of the SOR framework, this study identify discriminant categories of words to detect deceptive reviews. The experiment result showed that the accuracy of the proposed method can reach 80 percent.

中文摘要 ................................ ................................ ................................ ................................ ..... i
Abstract ................................ ................................ ................................ ................................ ....... ii
目錄 ................................ ................................ ................................ ................................ ........... iii
表目錄 ................................ ................................ ................................ ................................ ........ v
圖目錄 ................................ ................................ ................................ ................................ ....... vi
第一章 緒論 ................................ ................................ ................................ .............................. 1
1-1研究動機 ................................ ................................ ................................ ...................... 1
1-2研究目的 ................................ ................................ ................................ ...................... 2
1-3研究架構 ................................ ................................ ................................ ...................... 3
第二章 文獻探討 ................................ ................................ ................................ ...................... 4
2-1虛假評論 ................................ ................................ ................................ ...................... 4
2-2 S-O-R(Stimulus – Organism – Response)框架 ................................ ........................... 5
第三章 研究方法 ................................ ................................ ................................ ...................... 7
3-1 依據 S-O-R發展字詞類別 發展字詞類別 ................................ ................................ ........................ 7
3-2 挑選具有鑑別 力之類................................ ................................ ............................. 9
3-3 建立判斷法則 ................................ ................................ ................................ ............. 9
3-3-1計算字詞分數 .................................................................................................. 10
3-3-2 跨類別判斷法則 ............................................................................................. 10
3-4 判斷績效衡量方式 ................................ ................................ ................................ ... 11
第四章 研究實驗 ................................ ................................ ................................ .................... 13
4-1 資料集 ................................ ................................ ................................ ....................... 13
4-1-1 虛假評論資料來源 ......................................................................................... 13
4-1-2 真實評論資料來源 ......................................................................................... 13
4-2 挑選具鑑別 力的類................................ ................................ ............................... 14
4-3 從 Yelp 收集評論資料 收集評論資料 ................................ ................................ .......................... 21
iv
4-4 研究結果 ................................ ................................ ................................ ................. 21
4-4-1 領域與績效表 ................................................................................................. 21
4-4-2 評論數量與績效 ............................................................................................. 23
4-4-3 與其他方法的比較 ......................................................................................... 24
第五章 結論與建議 ................................ ................................ ................................ ................ 26
5-1 研究結論 ................................ ................................ ................................ ................... 26
5-2 未來研究建議 ................................ ................................ ................................ ........... 26
參考文獻 ................................ ................................ ................................ ................................ .. 28
附錄一 慾望字庫一 ................................ ................................ ................................ ................ 31
附錄二 慾望字庫二 ................................ ................................ ................................ ................ 34
附錄三 行動字庫 ................................ ................................ ................................ .................... 36

[1] Akoglu, L., Chandy, R., & Faloutsos, C. (2013). Opinion Fraud Detection in Online Reviews by Network Effects. ICWSM, 13, 2-11. [2] Baker, J. (1986). The role of the environment in marketing services: The consumer perspective. The services challenge: Integrating for competitive advantage, 1(1), 79-84. [3] Banerjee, S., & Chua, A. Y. (2014, September). Understanding the process of writing fake online reviews. In Digital Information Management (ICDIM), 2014 Ninth International Conference on (pp. 68-73). IEEE. [4] Bitner, M. J. (1992). Servicescapes: The impact of physical surroundings on customers and employees. The Journal of Marketing, 57-71. [5] Bohl, P. (2012). The effects of store atmosphere on shopping behaviour-A literature review. Corvinus Marketing Tanulmányok, 2012(1). [6] Bromley, R. D. F., & Thomas, C. T. (1995). The International Review of Retail, Distribution and Consumer Research. [7] Cresci, S., Di Pietro, R., Petrocchi, M., Spognardi, A., & Tesconi, M. (2015). Fame for sale: efficient detection of fake Twitter followers. Decision Support Systems, 80, 56-71. [8] Eisenstein, J., Ahmed, A., & Xing, E. P. (2011). Sparse additive generative models of text. [9] Eroglu, S. A., Machleit, K. A., & Davis, L. M. (2001). Atmospheric qualities of online retailing: A conceptual model and implications. Journal of Business research, 54(2), 177-184. [10] Eroglu, S. A., Machleit, K. A., & Davis, L. M. (2003). Empirical testing of a model of online store atmospherics and shopper responses. Psychology & Marketing, 20(2), 139-150. [11] Hirschberg, J., Benus, S., Brenier, J. M., Enos, F., Friedman, S., Gilman, S., ... & Pellom,
29
B. L. (2005, September). Distinguishing deceptive from non-deceptive speech. In Interspeech (pp. 1833-1836). [12] Hussain, R., & Ali, M. (2015). Effect of store atmosphere on consumer purchase intention. [13] Jacoby, J. (2002). Stimulus-organism-response reconsidered: An evolutionary step in modeling (consumer) behavior. Journal of Consumer Psychology, 12(1), 51-57. [14] Jeff, B. P. (2008). Consumer Behaviour Theory: Approaches and models. Unpublished discussion paper. http://eprints. bournemouth. ac. uk/101107. [15] Jindal, N., & Liu, B. (2007, October). Analyzing and detecting review spam. In Data Mining, 2007. ICDM 2007. Seventh IEEE International Conference on (pp. 547-552). IEEE. [16] Koufaris, M. (2002). Applying the technology acceptance model and flow theory to online consumer behavior. Information systems research, 13(2), 205-223. [17] Li, J., Ott, M., Cardie, C., & Hovy, E. H. (2014, June). Towards a General Rule for Identifying Deceptive Opinion Spam. In ACL (1) (pp. 1566-1576). [18] Manganari, E. E., Siomkos, G. J., & Vrechopoulos, A. P. (2009). Store atmosphere in web retailing. European Journal of Marketing, 43(9/10), 1140-1153. [19] Mihalcea, R., & Strapparava, C. (2009, August). The lie detector: Explorations in the automatic recognition of deceptive language. In Proceedings of the ACL-IJCNLP 2009 Conference Short Papers (pp. 309-312). Association for Computational Linguistics. [20] Mo, Z., Li, Y. F., & Fan, P. (2015). Effect of online reviews on consumer purchase behavior. Journal of Service Science and Management, 8(3), 419. [21] Mummalaneni, V. (2005). An empirical investigation of Web site characteristics, consumer emotional states and on-line shopping behaviors. Journal of Business Research, 58(4), 526-532. [22] Muruganantham, G., & Bhakat, R. S. (2013). A review of impulse buying
30
behavior. International Journal of Marketing Studies, 5(3), 149-160. [23] Ott, M., Choi, Y., Cardie, C., & Hancock, J. T. (2011, June). Finding deceptive opinion spam by any stretch of the imagination. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies-Volume 1 (pp. 309-319). Association for Computational Linguistics. [24] Pennebaker, J. W., Boyd, R. L., Jordan, K., & Blackburn, K. (2015). The development and psychometric properties of LIWC2015. [25] Scott, W. A. (1962). Cognitive complexity and cognitive flexibility. Sociometry, 405-414. [26] Sherman, E., Mathur, A., & Smith, R. B. (1997). Store environment and consumer purchase behavior: mediating role of consumer emotions. Psychology and Marketing, 14(4), 361-378. [27] Tausczik, Y. R., & Pennebaker, J. W. (2010). The psychological meaning of words: LIWC and computerized text analysis methods. Journal of language and social psychology, 29(1), 24-54. [28] Veena, D. (2005). A Model of Online Impulse Buying: An Empirical Study (Doctoral dissertation, Ph. D. dissertation, Washington State University). [29] Vieira, V. A. (2013). Stimuli–organism-response framework: A meta-analytic review in the store environment. Journal of Business Research, 66(9), 1420-1426. [30] 林美玉 . (2015). 識別攻擊評論之研究 識別攻擊評論之研究 -以旅館業為例 . 中央大學企業管理系位論 文, 1-36. [31] 張甜 . (2014). 辨別虛假評論之研 辨別虛假評論之研 究-以旅館業為例 . 中央大學企業管理系位論 文, 1-38. [32] 林佳勳 .(2015). 以謠言與謊理論為基礎之真假評識別研究 以謠言與謊理論為基礎之真假評識別研究 . 中央大學企業管理 學系位論 文,1-44