ShopGuard: A Multimodal Framework for Dark Pattern and Counterfeit Product Detection in E-Commerce

Author

Smita Mangesh Junnarkar, Hardik Poojary

Abstract

 Online shoppers increasingly face two intertwined threats: interface designs that manipulate purchasing decisions (dark patterns) and listings for counterfeit or fraudulent goods. Existing tools tend to address one threat at a time, rely on opaque learned scores, or overstate their conclusions. This paper presents ShopGuard, a multimodal trust-and-safety framework that analyses an e-commerce product URL, a product photograph, a product name, or a set of page screenshots, and produces a transparent 0–100 Trust Score together with quoted evidence for every finding. ShopGuard combines (i) a deterministic, rule-based text engine covering eight categories of dark-pattern and trust signals, (ii) a page-image module that inspects image provenance and reuse, (iii) a photo-based image check that uses only measurable signals and a vision-language model for product identification, (iv) OCR of screenshots so that text embedded in images passes through the same engine, and (v) live price comparison across verified Indian retailers, with an explicit “Price not available” outcome instead of estimated prices. The score is an interpretable penalty model whose every term can be traced to a specific finding. By design, ShopGuard reports suspicious signals and potential counterfeit indicators, never confirmed fraud, and it degrades gracefully to a screenshot-upload workflow when a site blocks automated access, without attempting to bypass security controls. In repository-level validation, 37/37 automated tests passed (100%) and all six built-in demos matched their predefined risk bands (6/6). A synthetic rule-coverage corpus yielded 96.88% macro Precision, 100.00% Recall and 98.21% F1, while a synthetic image-indicator regression achieved 100.00% Precision, Recall and F1; these are implementation-level regression results rather than independent benchmark measurements.

Keywords

Patterns, Deceptive Design, Counterfeit Detection, E-Commerce, Trust and Safety, Multimodal Analysis, OCR, Explainable Scoring, Price Comparison.

Full Text:

Download Paper PDF

References

[1] A. Mathur, G. Acar, M. J. Friedman, E. Lucherini, J. Mayer, M. Chetty, and A. Narayanan, “Dark patterns at scale: Findings from a crawl of 11K shopping websites,” Proc. ACM Hum.-Comput. Interact., vol. 3, no. CSCW, Art. 81, 2019.

[2] C. M. Gray, Y. Kou, B. Battles, J. Hoggatt, and A. L. Toombs, “The dark (patterns) side of UX design,” in Proc. CHI Conf. Human Factors in Computing Systems, 2018.

[3] A. Mathur, J. Mayer, and M. Kshirsagar, “What makes a dark pattern… dark? Design attributes, normative considerations, and measurement methods,” in Proc. CHI Conf. Human Factors in Computing Systems, 2021.

[4] J. Luguri and L. J. Strahilevitz, “Shining a light on dark patterns,” J. Legal Analysis, vol. 13, no. 1, pp. 43–109, 2021.

[5] L. Di Geronimo, L. Braz, E. Fregnan, F. Palomba, and A. Bacchelli, “UI dark patterns and where to find them: A study on mobile applications and user perception,” in Proc. CHI Conf. Human Factors in Computing Systems, 2020.

[6] Y. Yada, J. Feng, T. Matsumoto, N. Fukushima, F. Kido, and H. Yamana, “Dark patterns in e-commerce: A dataset and its baseline evaluations,” in Proc. IEEE Int. Conf. Big Data, 2022.

[7] S. M. Hasan Mansur, S. Salma, D. Awofisayo, and K. Moran, “AidUI: Toward automated recognition of dark patterns in user interfaces,” in Proc. IEEE/ACM Int. Conf. Software Engineering (ICSE), 2023.

[8] Central Consumer Protection Authority, Government of India, “Guidelines for prevention and regulation of dark patterns,” 2023.

[9] H. Brignull, “Deceptive design (formerly darkpatterns.org).” [Online]. Available: https://www.deceptive.design

[10] OECD and EUIPO, Trends in Trade in Counterfeit and Pirated Goods. Paris: OECD Publishing, 2019.

[11] A. Radford et al., “Learning transferable visual models from natural language supervision,” in Proc. Int. Conf. Machine Learning (ICML), PMLR vol. 139, 2021, pp. 8748–8763.

[12] Gemini Team, Google, “Gemini: A family of highly capable multimodal models,” arXiv:2312.11805, 2023.

[13] R. Smith, “An overview of the Tesseract OCR engine,” in Proc. Int. Conf. Document Analysis and Recognition (ICDAR), 2007, pp. 629–633.

Share your valuable work from Social Media Buttons