Project Stage 3

Matching Fodors and Zagats

• User ID: bugwriter
• Project ID: trial
• Screenshot:

Blocking Results

• User ID: bugwriter
• Project ID: movies
• Screenshot:

Matching Results

• User ID: bugwriter
• Project ID: movies
• Screenshot:

Estimating Accuracy

Links

• Prediction list
• Candidate set
• Table A
• Table B

Analysis

• Size of the candidate set C: 3179
• PDF file
• Blocking code: Because the program terminates at the first iteration, we don’t need any blocking rules/code to reduce the size of the candidate set C.
• Final reduced set of candidate tuple pairs: It is still the candidate set C.
• 400 tuples sampled and manually labeled
• Jupyter notebook output: ((0.989417801886328, 1.003202124312934), (1.0, 1.0))
• Recall = [0.989 - 1.003]
• Precision = [1.0 - 1.0]