WikiNex 📰📚

WikiNex is a full-stack web and mobile app, along with a Chrome extension, that enriches news articles with historical context using relevant excerpts from Wikipedia. The goal is to help users better understand the background and significance of current events.

🧑‍🎓 This project was built independently while I was in high school, as a way to explore natural language processing and historical relevance in journalism.

• 🔎 Automatically detects and processes news article text
• 🌐 Queries Wikipedia to find relevant historical background
• 🧠 Uses cosine similarity to extract the most relevant sections
• 📱 Available as:
  • Web App
  • Mobile App (React Native)
  • Chrome Extension
• 💾 Caching system for improved performance

• Backend: Python, Flask
• Frontend: HTML, CSS, JavaScript, Bootstrap, React Native
• APIs & NLP: OpenAI API, DPR (Dense Passage Retrieval) Transformer
• Other: Wikipedia API, cosine similarity algorithms, sentence tokenization

1. The app ingests a news article’s content.
2. It extracts relevant keywords and entities.
3. These are used to query Wikipedia articles.
4. Sentences from those articles are scored using cosine similarity.
5. The most relevant excerpts are shown as historical context.

Chrome Extension: https://docs.google.com/presentation/d/1ZWmAEuBnJqe8TDvK_DMxPRitG4MlR1nL09KOUS602uc/edit?usp=sharing

Mobile: https://github.com/user-attachments/assets/260be929-3fc2-4546-9058-2ed2e1c4b1f4