Detection of AI-Generated and Deceptive Fake Reviews on E-Commerce Platforms Using Lightweight Machine Learning | IJCT Volume 13 – Issue 6 | IJCT-V13I6P100

IJCT
International Journal of Computer Techniques
ISSN 2394-2231 · Peer-Reviewed · Open Access
📚 Volume 13, Issue 6
📅 November 19, 2026
📄 Pages 844–850
🔖 ID: IJCT-V13I6P100

Detection of AI-Generated and Deceptive Fake Reviews on E-Commerce Platforms Using Lightweight Machine Learning

Author(s)

Arpita Singh, Mahima Wankhade, Sanyogita Dahale, Snehal Patil

Abstract

Product reviews shape what people buy, yet their credibility is under a fresh threat: large language models (LLMs) can now mass-produce reviews that are fluent, varied and grammatically flawless. Current detectors cope with this only in part. The best-performing published systems lean on transformer models, paid LLM interfaces or reviewer metadata, resources that smaller platforms often lack, whereas lighter classical models were largely designed around older, human-written spam. We compare five recent studies of fake-review detection, looking at their data, labelling, methods, reported accuracy and acknowledged limitations. Two patterns emerge: accuracy tends to rise together with computational cost, and some of the highest accuracies rest on heuristic or weakly supervised labels, which casts doubt on what they actually measure. Drawing on these observations, we propose a lightweight, interpretable pipeline that pairs TF-IDF n gram features with stylometric indicators of machine generated text and compares four classifiers: Naïve Bayes, Logistic Regression, Support Vector Machine and Random Forest. The accompanying evaluation protocol is fully specified and includes stratified hold-out and cross validation, cross-generator and cross-domain tests, paraphrase robustness and ablation. Experimental results for the proposed pipeline are not included in this version, and no performance figures are claimed for it. The system is intended as a risk-scoring aid for human moderators, never as proof of AI authorship.

Keywords

AI-generated reviews, fake review detection, opinion spam, large language models, TF-IDF, stylometry, Random Forest, e-commerce trust.

Conclusion

Examining recent machine-learning and deep-learning studies of fake and AI-generated review detection revealed three consistent patterns. Accuracy improves along with computing cost and metadata requirements; the most striking accuracy figures depend on label construction that may favour the model; and few studies test generalisation to new generators, domains or paraphrased text. In response, the paper proposed a lightweight, interpretable pipeline that combines TF-IDF n-grams with stylometric features, compares Naïve Bayes, Logistic Regression, SVM and Random Forest, and outputs a calibrated risk score. An evaluation protocol designed to avoid the identified validity problems accompanies it. The pipeline is a screening aid for human moderators and is not proof that a review was machine-written.

References

[1] S. He, B. Hollenbeck, and D. Proserpio, "The market for fake reviews," Marketing Sci., vol. 41, no. 5, pp. 896–921, 2022.
[2] N. Jindal and B. Liu, "Opinion spam and analysis," in Proc. Int. Conf. Web Search Web Data Mining (WSDM), 2008, pp. 219–230.
[3] M. Ott, Y. Choi, C. Cardie, and J. T. Hancock, "Finding deceptive opinion spam by any stretch of the imagination," in Proc. 49th Annu. Meeting Assoc. Comput. Linguistics (ACL-HLT), 2011, pp. 309–319.
[4] W. Liao et al., "Differentiate ChatGPT-generated and human-written medical texts," arXiv:2304.11567, 2023.
[5] H. Alamleh, A. A. S. AlQahtani, and A. ElSaid, "Distinguishing human written and ChatGPT-generated text using machine learning," in Proc. Syst. Inf. Eng. Design Symp. (SIEDS), 2023, pp. 154–158.
[6] I. Katib, F. Y. Assiri, H. A. Abdushkour, D. Hamed, and M. Ragab, "Differentiating Chat Generative Pretrained Transformer from humans: Detecting ChatGPT-generated text and human text using machine learning," Mathematics, vol. 11, no. 15, Art. no. 3400, 2023.
[7] C. U. Rao, Y. L. Lahari, V. Nitin, K. Shashank, B. Mahesh, and B. Anil, "A multi-factor AI-driven framework for detecting fake and AI-generated online reviews using semantic, behavioral, and temporal analysis," in Proc. 6th Int. Conf. Expert Clouds Appl. (ICOECA), 2026, pp. 1563 1570, doi: 10.1109/ICOECA68095.2026.11485495.
[8] V. Khanna M A, K. Ramalakshmi, D. P. P. Latha, R. Venkatesan, and A. C. Wanika R, "An efficient approach for fake review detection with machine and deep learning techniques," in Proc. 7th Int. Conf. Innov. Data Commun. Technol. Appl. (ICIDCA), 2025, pp. 1012–1016, doi: 10.1109/ICIDCA66325.2025.11280411.
[9] S. L. Sajja, R. Pranavi, T. N. Amulya, S. S. Priya, Y. Aparna, and S. Madhumitha, "Enhanced fake review detection in e-commerce: A comparative study of deep learning and hybrid ensemble classifiers," in Proc. 4th Int. Conf. Augmented Intell. Sustain. Syst. (ICAISS), 2026, pp. 1589–1593, doi: 10.1109/ICAISS68683.2026.11525550.
[10] T. Pasupathi Pandi and N. Siva Kumar, "Fake review detection in e commerce using machine learning and NLP technique," in Proc. 3rd Int. Conf. Inventive Comput. Informatics (ICICI), 2025, pp. 692–698, doi: 10.1109/ICICI65870.2025.11069636. • Embeddings and transformers. Test word embeddings, LSTM and fine-tuned DistilBERT or RoBERTa as optional higher-accuracy tiers, including distillation into the lightweight model. • Behavioural and temporal signals. Where a platform shares them, combine text with reviewer behaviour and posting bursts, as earlier multi-factor work has done
[7][17]. • Adversarial and multilingual evaluation. Study adaptive adversaries who paraphrase detected text, and extend the work to other languages and to non-text signals. • Deployment. Build a moderator dashboard with explanations, and study how analysts use the risk bands. ACKNOWLEDGMENT The authors thank their guide, Mrs. Snehal Patil, and Dr. D. Y. Patil Institute of MCA and Management, Akurdi, Pune, for guidance and support.
[11] L. Bankadamani, V. R. Honnagudi, B. Gudi, S. Bhagoji, C. Muttal, and G. Konnurmath, "Fake review detection in fashion e-commerce using machine learning," in Proc. IEEE Int. Conf. Convergence Comput. Technol. (I3CTCON), Lonavala, 10.1109/I3CTCON68242.2026.11507156. India, 2026, doi:
[12] S. Gehrmann, H. Strobelt, and A. M. Rush, "GLTR: Statistical detection and visualization of generated text," arXiv:1906.04043, 2019.
[13] E. Mitchell, Y. Lee, A. Khazatsky, C. D. Manning, and C. Finn, "DetectGPT: Zero-shot machine-generated text detection using probability curvature," arXiv:2301.11305, 2023.
[14] V. Sanh, L. Debut, J. Chaumond, and T. Wolf, "DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter," arXiv:1910.01108, 2019.
[15] S. Geetha, E. Elakiya, R. Sujithra Kanmani, and M. K. Das, "High performance fake review detection using pretrained DeBERTa optimized with monarch butterfly paradigm," Sci. Rep., Mar. 2025.
[16] R. Mohawesh, H. B. Salameh, Y. Jararweh, M. Alkhalaileh, and S. Maqsood, "Fake review detection using transformer-based enhanced LSTM and RoBERTa," Array, 2024.
[17] D. Zhang, L. Zhou, J. L. Kehoe, and I. Y. Kilic, "What online reviewer behaviors really matter? Effects of verbal and nonverbal behaviors on detection of fake online reviews," MIS Quart., 2016.
[18] J. Salminen, C. Kandpal, A. M. Kamel, S. Jung, and B. J. Jansen, "Creating and detecting fake reviews of online products," Decis. Support Syst., 2022.
[19] W. Liang, M. Yuksekgonul, Y. Mao, E. Wu, and J. Zou, "GPT detectors are biased against non-native English writers," Patterns, vol. 4, no. 7, Art. no. 100779, 2023.
[20] C. Cortes and V. Vapnik, "Support-vector networks," Mach. Learn., vol. 20, no. 3, pp. 273–297, 1995.
[21] L. Breiman, "Random forests," Mach. Learn., vol. 45, no. 1, pp. 5–32, 2001.
[22] F. Pedregosa et al., "Scikit-learn: Machine learning in Python," J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.
[23] S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," in Adv. Neural Inf. Process. Syst. (NeurIPS), vol. 30, 2017.

📋 How to Cite This Paper

Arpita Singh, Mahima Wankhade, Sanyogita Dahale, Snehal Patil (2026). Detection of AI-Generated and Deceptive Fake Reviews on E-Commerce Platforms Using Lightweight Machine Learning. International Journal of Computer Techniques, 13(6), 844–850. ISSN: 2394-2231. DOI: https://doi.org/10.5281/zenodo.24370423
© 2026 International Journal of Computer Techniques (IJCT). All rights reserved. · ijctjournal.org
Submit Your Paper