Skip to content

Latest commit

History

History
42 lines (31 loc) 路 1.41 KB

File metadata and controls

42 lines (31 loc) 路 1.41 KB

Team Name: [Insert Team Name]

馃懃 Team Members

  • Member 1: [Name] - [GitHub Profile]
  • Member 2: [Name] - [GitHub Profile]
  • Member 3: [Name] - [GitHub Profile]
  • Member 4: [Name] - [GitHub Profile]

馃敆 Project Links

  • Demo Video: [Link to YouTube/Loom (3-5 mins)]
  • Hosted Demo: [Optional: Link to live app]

馃 Technical Implementation

1. Data Pipeline (Scraping)

  • Tools Used: (e.g., Scrapy, Beautiful Soup, Selenium)
  • Strategy: Briefly explain how you scraped wiki.metakgp.org. How did you handle cleaning and indexing?
  • Indexing: (e.g., ChromaDB, FAISS, Pinecone)

2. Graph of Thoughts (GoT)

  • Reasoning Model: Explain how your nodes and edges are structured.
  • Graph Logic: How do you connect different MetaKGP pages? (e.g., connecting a "Society" page to a "Student" page).

3. Mixture of Experts (MoE)

  • Expert 1 (Source Matcher): How do you verify the text exists in the scraped data?
  • Expert 2 (Hallucination Hunter): How do you detect fabricated info?
  • Expert 3 (Logic Expert): How do you ensure logical consistency?

馃搳 Setup Instructions

  • Prerequisites: (e.g., Python 3.10, Neo4j)
  • Environment Variables: List required .env keys (do not share actual keys).
  • How to Run Scraper: python scrape.py
  • How to Run Bot: streamlit run app.py

馃摳 Screenshots

[Image of Graph Visualization]

[Image of Chat Interface]