Tackling Challenge 2: Digital integrity in the age of misinformation
Empowering informed conversations: AI-powered sentiment & bias analysis on Malaysia’s news
Check out the live demo of Tea Time: 👉 Click here to try it out
We noticed even perfectly accurate information can get twisted as it passes from person to person, just like the classic “telephone game.” For instance, the word “Polish” might be taken as the country or as nail cleaner, and “present” can mean now or gift, depending on who hears it. Misinterpretation often happens not because the facts are wrong, but because people miss the broader context.
“What if there were a way to let everyone see the full picture, so that every nuance of a news article is preserved?”
That question sparked Tea Time. By combining crowdsourced commentary with AI-driven sentiment analysis (to see how everyone comments) and bias analysis (to flag where an article slants), then providing a concise summary to guide interpretation, we ensure no detail is lost in transmission and that misinformation gets caught before it spreads.
This exploration led us to design the platform Tea Time as a responsive PWA with support for any device, tailored for Malaysia, that:
- Aggregates news directly from Bernama and other sources
- Allows anyone to comment on any article, thus building a crowdsourced view of public sentiment
- Uses AI to classify comments as positive, negative, or neutral
- Generates bias-analysis reports on each article, flagging potential slants and offering reading guidance
- Highlights “the bigger picture” to prevent misinterpretation by using AI
- Implements KYC verification to ensure user accountability and constructive engagement
- Encrypts user data using elliptic curve cryptography for secure storage in the database
- Leverages OpenAI to fetch and suggest related news and context, helping users explore more about the topic
Clone the repository and start the development server:
git clone https://github.com/JingYuan0926/tt.git
cd tt
npm install
npm run dev
- Cursor for code assistant
- ChatGPT for code debugging
- NextUI for frontend components
- shadcn/ui for UI primitives
- HeroUI for design elements
- Tailwind CSS for utility-first styling
- GNews API for news aggregation
- Resend API for OTP delivery
- Stripe for payment processing
- MongoDB as the primary database
- OpenAI GPT-4o-mini for AI-driven sentiment & bias analysis & AI summaries
- Gemini 1.5 flash for document parsing & KYC
- Elliptic Curve Cryptography for secure KYC data encryption
- Google Chrome Extension for in-page AI summaries & bias checks
- Progressive Web App (PWA) support for any device including web and mobile
-
NLP for Comments Processing
/pages/api/analyzeComments.js -
Computer Vision for Document Parsing
/pages/api/parse-ic-documents.js -
Multi-Platform Support
/extension -
Stripe Payment for Secure Checkout
/pages/api/create-checkout-session.js -
Elliptic Curve Cryptography encryption
/lib/cryptography.js -
MongoDB Database Storage
/pages/api/comments.js
/pages/api/fetch-news.js
/pages/api/register.js -
LLM for News Analysis and Recommendation
/pages/api/analyze-news.js -
AI for Multi-News Source Aggregation
/pages/api/generate-sources.js
A browser extension powered by OpenAI to enhance user productivity with intelligent features same as our platform, just a lightweight version that will work anywhere, anytime!
Follow these steps to set up and run the extension locally.
Clone the project from GitHub:
git clone https://github.com/JingYuan0926/tt.git
cd ttnpm installCreate a .env file in the root directory and add the following:
OPENAI_API_KEY=your_openai_api_key_here
- Run the Development Server
Start the extension in development mode:
npm run dev
Note: The web application can be accessed via the deployed link, this setup is to run the web extension only!