Skip to content

This repository for team "it_works_on_local" participating in the "Dark Pattern Buster Hackathon"!

Notifications You must be signed in to change notification settings

ajaman190/BrightBrowser

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

29 Commits
 
 
 
 
 
 

Repository files navigation

DarkPatternBuster: BrightBrowse Web Extension

Table of Contents

  1. Introduction
  2. Features Overview
  3. Dark Pattern Detection Models
  4. Further Reading and Links
  5. Team it_works_on_local

Introduction

BrightBrowse is a groundbreaking web browser extension designed to empower users with the ability to detect, understand, and navigate through the maze of dark patterns commonly employed by e-commerce platforms. This solution will promote transparency and ethical practices in online shopping, fostering a safer and more informed user experience.


Features Overview

BrightBrowse is equipped with a suite of features aimed at enhancing user security and awareness:

Feature Description Benefit
User Authentication Secure access through manual ID/password or Google sign-in. Personalized and secure user experience.
Pattern Detection Identification and solutions for various dark patterns. Safer browsing by avoiding deceptive tactics.
Personalized Settings Customization options to tailor extension functionality. Enhanced user experience aligned with individual preferences.
Education Regular updates with tips, articles, and blogs on dark patterns. Increased awareness and knowledge about deceptive practices.
Reporting Users can report new dark patterns, contributing to the community. Empowers users to contribute and stay engaged.
History & Insights Tracks and displays the history of encountered patterns and user feedback. Personalized data for better understanding and improvement.

Dark Pattern Detection Models

We integrated advanced models to detect a variety of dark patterns, including:

  • Deceptive UI Detection: Utilizing logistic regression to identify misleading user interfaces.
  • Fake Review Detection: A RoBERTa-based model distinguishing genuine from fake reviews.
  • Malicious URL Detection: An XGBoost classifier to categorize URLs and protect users from harmful links.
  • Price Manipulation Detection: Analyzing price trends to identify potential manipulative practices.
  • Basket Sneaking Detection: Uses YOLO to spot unapproved cart additions via price text analysis.
  • Trick Question Detection: Applies keyword scoring to reveal trick questions.
  • Countdown Timer Detection: Flags countdowns through text-based numerical decline.
  • Forced Continuity Detection: Detects hidden fees in English using specific regex.

Further Reading and Links

  • Extension README - Delve into the technical aspects of our browser extension.
  • Backend README - Explore the backend architecture and technologies powering BrightBrowse.

Team it_works_on_local

We're a dynamic team of pre-final year undergraduate students at IIT Kharagpur, specializing in development and data science.


BrightBrowse - Illuminating the Shadows of Dark Patterns.

About

This repository for team "it_works_on_local" participating in the "Dark Pattern Buster Hackathon"!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published