Data Science Project - CS 839

Team Members

• Sri Harshal Parimi ([email protected])
• Shebin Roy Yesudhas ([email protected])
• Sankarshan Umesh Bhat ([email protected])

Stage-1: Information Extraction from natural text

• Set B
  • README
• Set I
• Set J
• Source code
• Compressed file
• Stage 1 Report

Stage-2: Crawling and Extracting Structured data from web-pages

• Data
• Source code
• Report

Stage-3: Entity-matching

Matching Fodors and Zagats

• UserId: Avengers
• ProjectId: endgame
• Screenshot

Blocking Results

• UserId: Avengers
• ProjectId: MoviesMatcher
• Screenshot

Matching Results

• UserId: Avengers
• ProjectId: MoviesMatcher
• Screenshot

Estimating accuracy

• Candidate set - 72165 tuple pairs

• Prediction list

• Table A

• Table B

• Candidate set size is 72165 which is greater than 500

• Report for blocking rules

• Code for blocking rules

• Reduced candidate set

• Labeled Tuple pairs

• Recall = [0.9371096866388409 - 0.9910340259360095]

• Precision = [0.9186143717366582 - 0.9782780006778256]