Deep Think Project

This project provides the frontend interface for the Deep Think service — a critical reading tool that uses AI to analyze and classify sentences from articles as either clear or ambiguous. It enables users to easily interact with the underlying analysis engine through an intuitive web-based experience.

---

🚀 Features

• Accepts a URL or copied text via the input search bar
• Visualizes AI-powered classification results in two types:
  • Clear sentence: Objectively interpretable or fact-based
  • Ambiguous sentence: Subjective, emotional, or open to multiple interpretations
• Offers sentence-level tooltips to explain each classification
• Provides summary and author intent analysis after sentence classification
• Links ambiguous sentences to external references for deeper insight
• Fully responsive user interface with smooth transitions and tooltip overlays

---

🧭 How to Use

🏠 Main Page

Copy a news article URL or drag-select and copy text.
Paste it into the search bar, then click the Analyze button.

---

⏳ Loading Page

While processing, Deep Think analyzes each sentence and classifies it into one of two categories.

---

📊 Result Page

• Summary & Author Intent:
  A brief summary and author perspective are displayed at the top.

  

• Highlighted Sentences:
  The full text is annotated using colored highlights.
  Green = clear, Purple = ambiguous.

  

• Interactive Tooltips:
  Clicking on a sentence reveals additional insights:

  • For clear (green) sentences:
    • The classification rationale

  

  • For ambiguous (purple) sentences:
    • The classification rationale
    • Additional interpretations
    • Related reference links (when available)

  

• Reference Material:
  External links open in a new tab and provide supporting sources for ambiguous sentences.

---

⚙️ Setup Instructions

To run this frontend locally:

1. Set your backend proxy URL In the project root, open your package.json file and set the proxy field to your backend server URL

   // package.json
   {
     ...
     "proxy": "http://your_proxy_address"
   }

2. Install dependencies

   npm install
   

3. Start the development server

   npm start
   

This will launch the app at http://localhost:3000 by default.

---

🧑‍💻 Author

This frontend was built as part of the Deep Think project, an initiative to promote media literacy and critical thinking by leveraging generative AI. The UI is designed for simplicity, speed, and interpretability — helping users better understand and question the information they consume.

---

📜 License

MIT License