Detection & Classification of Spam Reviews and Spammer Communities in E-Commerce

The rapid expansion of e-commerce platforms has made online reviews a crucial factor influencing consumer purchasing decisions. However, the increasing presence of spam and deceptive reviews poses significant challenges to trust, transparency, and fair market competition. Fake reviews mislead customers, damage brand reputation, and distort product rankings, highlighting the urgent need for reliable detection mechanisms. This review examined various approaches for detecting spam reviews, including linguistic analysis, behavioral modeling, machine learning classifiers, and deep learning techniques. Traditional methods such as Naïve Bayes, Support Vector Machines, and Decision Trees provide efficient and interpretable solutions, while advanced deep learning and network-based approaches enhance detection accuracy by capturing contextual patterns and coordinated spam behavior. Feature extraction techniques and reviewer behavior analysis further improve classification performance. Despite significant progress, challenges remain due to evolving spam strategies, limited labeled datasets, and computational complexity. Future research should focus on hybrid, scalable, and explainable models capable of real-time detection and spammer community identification. Implementing such intelligent systems will improve consumer trust, support fair competition, and strengthen the reliability of e-commerce platforms.

[1] N. Jindal and B. Liu, “Opinion spam and analysis,” ACM SIGKDD Explorations Newsletter, vol. 10, no. 2, pp. 219–230, 2008. [2] S. Rayana and L. Akoglu, “Collective opinion spam detection using network effects,” ACM Transactions on Intelligent Systems and Technology, vol. 8, no. 2, pp. 1–27, 2016. [3] M. Ott, Y. Choi, C. Cardie, and J. T. Hancock, “Finding deceptive opinion spam by any stretch of the imagination,” in Proc. ACL, 2011, pp. 309–319. [4] H. Li, Z. Chen, B. Liu, X. Wei, and J. Shao, “Spotting fake reviews via collective positive-unlabeled learning,” in Proc. IEEE Int. Conf. Data Mining (ICDM), 2014, pp. 899–904. [5] C. Xu and J. Zhang, “Combating product review spam campaigns via multiple heterogeneous pairwise features,” in Proc. SIAM Int. Conf. Data Mining (SDM), 2015, pp. 172–180. [6] S. Mukherjee, A. Kumar, B. Liu, J. Wang, M. Hsu, and M. Castellanos, “Spotting opinion spammers using behavioral footprints,” in Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., 2013, pp. 632–640. [7] H. Wang, Y. Lu, and C. Zhai, “Latent aspect rating analysis for review spam detection,” IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 3, pp. 454–468, 2020. [8] Y. Ren and Y. Ji, “Neural network-based detection of fake online reviews,” Expert Systems with Applications, vol. 168, Art. no. 114318, 2021. [9] X. Li, Z. Chen, and L. Huang, “BERT-based fake review detection in e-commerce platforms,” Information Processing & Management, vol. 59, no. 2, 2022. [10] P. Sharma and R. Mehta, “Explainable artificial intelligence for fake review detection in online marketplaces,” IEEE Access, vol. 12, pp. 14567–14579, 2024.