CS 4250 – Assignment 2 Summary

Course: Web Search and Recommender Systems
Assignment Title: Retrieval Models & PageRank
Due Date: May 9, 2025
Type: Group Project (same groups as before)

---

🔧 Project Parts

Part 1: Crawling and Indexing

• Perform a new crawl of at least 500 pages.
• Create a simple inverted index to map terms to documents.

Part 2: Simple Retrieval

• Implement a basic Boolean retrieval system.
  • No need to support complex operators (e.g., OR, NOT).
  • Input queries via command line:

    > Please enter your query: tropical fish  
    > Relevant results are: doc12, doc24, doc45
    

  • No GUI required.

Part 3: Improved Retrieval

• Implement an advanced retrieval model, such as:
  • Ranked Boolean
  • Vector Space Model
  • BM25

Part 4: PageRank

• Compute PageRank scores for the pages you crawled.
• You may need to recrawl to ensure enough link data.

Part 5: Combine

• Combine your retrieval model with PageRank.
  • Example: Multiply retrieval score × PageRank score.

---

📦 Deliverables

1. Code

   • All source code used to implement the project.

2. Report (PDF, max 5 pages) including:

   • Main components of the system.
   • Challenges faced during development.
   • Top 100 most important pages (by PageRank), with their scores.
   • Search results and performance comparison for 3 queries using different models.
   • Discussion on how PageRank influenced the rankings.
   • Contributions from each team member.

---

📅 Deadline

Submit by: May 9, 2025