CoDark: Navigate the Web with Transparency and Trust Combat deceptive website design and empower your online experience with CoDark, a powerful Chrome extension.

Developed by:

1. Shaurya Vardhan Mishra
2. Yatendra Pachori
3. Yagyadeep Ahirwar
4. Vivek Vaibhav
5. Nihal Patidar

Protect Yourself from Dark Patterns:

1. Advanced Language Model (LLM) Detection: CoDark harnesses the power of LLM technology, including XLNet and BERT, to identify and highlight manipulative design elements ("dark patterns") employed by websites.
2. Real-time Alerts: Be instantly notified upon encountering potential dark patterns, ensuring you browse with full awareness.
3. Comprehensive Dataset: Trained on a curated dataset, CoDark boasts accurate and nuanced pattern detection, safeguarding you from a wide range of deceptive tactics.

Embrace Transparency and User Choice:

1. Automated Page Analysis: Upon visiting any website, CoDark seamlessly analyzes the page in real-time, illuminating potential dark patterns.
2. Granular Highlighting: Suspicious elements are highlighted directly on the page, drawing your attention to manipulative design practices.
3. Deepen Your Understanding: Access educational resources within the extension to learn about dark patterns and equip yourself with the knowledge to make informed online decisions.
4. Contribute to a More Ethical Web.
5. Share Your Feedback: CoDark is continuously evolving. Share your experience, suggestions, and ideas to shape the future of the extension and contribute to a safer online environment through our website.

Contact the Developers: Reach out to the CoDark team at [[email protected]] with any feedback, questions, or collaboration opportunities.

Technical Specifications:

Development Framework: JavaScript (JS) Data Preprocessing and Model Training: Python libraries (e.g., Pandas, Torch, scikit-learn) Extension Deployment: Chromium and Web Extension API Dataset Sources: Publicly available datasets curated from relevant research papers and workshops

Inspired by: "Dark Patterns at Scale: Findings from a Crawl of 11K Shopping Websites" by Mathur et al. IEEE BigData 2022 Workshop Dataset by YadaYuki

Disclaimer: While inspired by these works, CoDark's dataset and implementation are unique and developed independently by the Jabalpur Engineering College team. Together, let's build a web space where transparency and user empowerment reign supreme. Download CoDark today and experience the difference!