Detecting Deceptive Dark Patterns in E-commerce Platforms

TL;DR

This paper addresses the challenge of detecting deceptive UI practices (dark patterns) on e-commerce sites. It proposes a pipeline that combines web scraping (BeautifulSoup4 and Selenium) with a fine-tuned BERT model to detect and classify dark patterns at the line level, using an eight-category taxonomy plus Not Dark Pattern. A manually annotated dataset from diverse sites enables supervised learning, achieving ~96% test accuracy and enabling threshold-based flagging of potential dark patterns. The study situates its approach within a broad literature base and demonstrates practical scalability for consumer protection, awareness, and regulatory monitoring in online shopping. Overall, the work offers a concrete, scalable framework for auditing e-commerce interfaces for deceptive design practices.

Abstract

Dark patterns are deceptive user interfaces employed by e-commerce websites to manipulate user's behavior in a way that benefits the website, often unethically. This study investigates the detection of such dark patterns. Existing solutions include UIGuard, which uses computer vision and natural language processing, and approaches that categorize dark patterns based on detectability or utilize machine learning models trained on datasets. We propose combining web scraping techniques with fine-tuned BERT language models and generative capabilities to identify dark patterns, including outliers. The approach scrapes textual content, feeds it into the BERT model for detection, and leverages BERT's bidirectional analysis and generation abilities. The study builds upon research on automatically detecting and explaining dark patterns, aiming to raise awareness and protect consumers.

Figures (3)

• Figure 1: Pending Model Architecture
• Figure 2: Famous e-commerce website 1
• Figure 3: Famous e-commerce website 2