WF-CFRB: A Deep Learning Approach for Fake Review Detection Based on Weighted Fusion of Contextual Features and Reviewer Behaviors

Abstract

Due to the increasing importance of online product reviews, how to accurately identify fake reviews has become an issue of concern to enterprises and consumers. The contextual features encapsulate the semantic information of review, while the behavioral features reflect the behavioral patterns of reviewers. However, an appropriate method to integrate contextual and behavioral features is a challenging task, hence an end-to-end model based on Weighted Fusion of Contextual Features and Reviewer Behaviors (WF-CFRB) for fake review detection is proposed. Firstly, the categories of average cosine similarity and the corpus of review are jointly fed into BERT to obtain contextual feature vectors. Then, the underlying patterns of the reviewer behaviors are extracted by CNN to construct behavioral feature vectors. Finally, a weighted fusion method is adopted to fuse contextual and behavior features for fake review detection. WF-CFRB and each component are evaluated on YELP dataset. WF-CFRB achieves F1 score of 81.31% and AUC score of 81.27%, and it also outperforms the other baseline models in terms of accuracy and recall. Compared with the original BERT model, the experimental results indicate that cosine similarity provides BERT with more information, which is useful to construct the contextual feature vectors. Through the weighted fusion of contextual and behavioral features, WF-CFRB yields excellent performance on fake review detection, which is particularly suitable for scenarios where behavioral features can be captured.

Keywords

Fake review detection / BERT / contextual features / reviewer behaviors / weighted fusion

References

[1] Aghakhani H, Machiry A, Nilizadeh S. et al.. Detecting deceptive reviews using generative adversarial networks. 2018 IEEE Security and Privacy Workshops (SPW), 201889-95.

[2] Banerjee S, Chua A Y K, Kim J J. Don’t be deceived: Using linguistic analysis to learn how to discern online review authenticity. Journal of the Association for Information Science and Technology, 2017, 68(6): 1525-1538.

[3] Budhi G S, Chiong R, Wang Z. et al.. Using a hybrid content-based and behaviour-based featuring approach in a parallel environment to detect fake reviews. Electronic Commerce Research and Applications, 2021, 47: 101048.

[4] Burgoon J K. Rejoinder to Levine, Clare et al.’s comparison of the Park-Levine probability model versus interpersonal deception theory: Application to deception detection. Human Communication Research, 2015, 41(3): 327-349.

[5] De Caigny A, Coussement K, De Bock K W. A new hybrid classification algorithm for customer churn prediction based on logistic regression and decision trees. European Journal of Operational Research, 2018, 269(2): 760-772.

[6] Devlin J, Chang M W, Lee K, et al. (2018). BERT: Pretraining of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.

[7] Duma R A, Niu Z, Nyamawe A S. et al.. A Deep Hybrid Model for fake review detection by jointly leveraging review text, overall ratings, and aspect ratings. Soft Computing, 2023, 27(10): 6281-6296.

[8] Fei G, Mukherjee A, Liu B. et al.. Exploiting burstiness in reviews for review spammer detection. Proceedings of the International AAAI Conference on Web and Social Media, 2013, 7(1): 175-184.

[9] Hussain N, Mirza H T, Hussain I. et al.. Spam review detection using the linguistic and spammer behavioral methods. IEEE Access, 2020, 8: 53801-53816.

[10] Javed M S, Majeed H, Mujtaba H, et al. (2021). Fake reviews classification using deep learning ensemble of shallow convolutions. Journal of Computational Social Science: 1–20.

[11] Jindal N, Liu B. Analyzing and detecting review spam. Seventh IEEE International Conference on Data Mining (ICDM 2007), 2007547-552.

[12] Jindal N, Liu B. Opinion spam and analysis. Proceedings ofthe 2008 International Conference on Web Search and Data Mining, 2008219-230

[13] Lan Z, Chen M, Goodman S, et al. (2019). Albert: A lite bert for self-supervised learning of language representations. arXiv preprint arXiv:1909.11942.

[14] Li F H, Huang M, Yang Y. et al.. Learning to identify review spam. Proceedings of the Twenty-Second international joint conference on Artificial Intelligence-Volume Volume Three, 20112488-2493

[15] Liu Y, Ott M, Goyal N, et al. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv preprintarXiv:1907.11692.

[16] Lozano M G, Brynielsson J, Franke U. et al.. Veracity assessment of online data. Decision Support Systems, 2020, 129: 113132.

[17] Ma M, Agarwal R. Through a glass darkly: Information technology design, identity verification, and knowledge contribution in online communities. Information Systems Research, 2007, 18(1): 42-67.

[18] Manaskasemsak B, Tantisuwankul J, Rungsawang A (2021). Fake review and reviewer detection through behavioral graph partitioning integrating deep neural network. Neural Computing and Applications: 1–14.

[19] Mikolov T, Chen K, Corrado G, et al. (2013). Efficient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781.

[20] Mohawesh R, Xu S, Tran S N. et al.. Fake reviews detection: A survey. IEEE Access, 2021, 9: 65771-65802.

[21] Mukherjee A, Kumar A, Liu B. et al.. Spotting opinion spammers usingbehavioral footprints. Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2013632-640.

[22] Mukherjee A, Venkataraman V, Liu B. et al.. What yelp fake review filter might be doing?. Proceedings of the International AAAI Conference on Web and Social Media, 2013, 7(1): 409-418.

[23] Murphy RLocal Consumer Review Survey 2023, 2023

[24] Noekhah S, binti Salim N, Zakaria N H. Opinion spam detection: Using multi-iterative graph-based model. Information Processing & Management, 2020, 57(1): 102140.

[25] Pennington J, Socher R, Manning C D. Glove: Global vectors for word representation. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 20141532-1543.

[26] Quinlan J R, Rivest R L. Inferring decision trees using the minimum description length principle. Information and Computation, 1989, 80(3): 227-248.

[27] Ren Y, Zhang Y. Deceptive opinion spam detection using neural network. Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, 2016140-150

[28] Sahut J M, Hajek P. Mining behavioural and sentiment-dependent linguistic patterns from restaurant reviews for fake review detection. Technological Forecasting and Social Change, 2022, 177(C): 12532

[29] Saidani N, Adi K, Allili M S. A semantic-based classification approach for an enhanced spam detection. Computers & Security, 2020, 94: 101716.

[30] Sanh V, Debut L, Chaumond J, et al. (2019). DistilBERT, a distilled version of BERT: Smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.

[31] Shan G, Zhou L, Zhang D. From conflicts and confusion to doubts: Examining review inconsistency for fake review detection. Decision Support Systems, 2021, 144: 113513.

[32] Wang C C, Day M Y, Chen C C. et al.. Detecting spamming reviews using long short-term memory recurrent neural network framework. Proceedings of the 2nd International Conference on E-commerce, E-Business and E-Government, 201816-20.

[33] Wang Q, Zhang W, Li J. et al.. Benefits or harms? The effect of online review manipulation on sales. Electronic Commerce Research and Applications, 2023, 57: 101224.

[34] Wolf T, Debut L, Sanh V. et al.. Transformers: State-of-the-art natural language processing. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, 202038-45.

[35] Xiong Y, Feng Y, Wu H. et al.. Fusing label embedding into BERT: An efficient improvement for text classification. Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, 20211743-1750.

[36] Ye J, Akoglu L. Discovering opinion spammer groups by network footprints. Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2015, Porto, Portugal, September 7–11, 2015, Proceedings, Part I 15, 2015267-282.

[37] Zhang D, Li W, Niu B. et al.. A deep learning approach for detecting fake reviewers: Exploiting reviewing behavior and textual information. Decision Support Systems, 2023, 166: 113911.

[38] Zhang D, Zhou L, Kehoe J L. et al.. What online reviewer behaviors really matter? Effects of verbal and nonverbal behaviors on detection of fake online reviews. Journal of Management Information Systems, 2016, 33(2): 456-481.

[39] Zhang L, Ma B, Cartwright D K. The impact of online user reviews on cameras sales. European Journal of Marketing, 2013, 47(7): 1115-1128.

[40] Zhang L, Wu Z, Cao J. Detecting spammer groups from product reviews: A partially supervised learning model. IEEE Access, 2017, 6: 2559-2568.

[41] Zhang W, Xie R, Wang Q. et al.. A novel approach for fraudulent reviewer detection based on weighted topic modelling and nearest neighbors with asymmetric KullbackLeibler divergence. Decision Support Systems, 2022, 157: 113765.

[42] Zhang W, Zhang X, Chen J, et al. (2024). Stacking GA2M for inherently interpretable fraudulent reviewer identification by fusing target and non-target features. International Journal of General Systems: 1–36.

[43] Zhang Z, Han X, Liu Z, et al. (2019). ERNIE: Enhanced language representation with informative entities. arXiv preprint arXiv:1905.07129.

[44] Zhou L, Burgoon J K, Twitchell D P, et al. (2004). A comparison of classification methods for predicting deception in computer-mediated communication. Journal of Management Information Systems: 139–165.

RIGHTS & PERMISSIONS

Systems Engineering Society of China and Springer-Verlag GmbH Germany