Project: Data Matching
Topic: Restaurants
Team: Clarence Cheung, Jin Ruan
Description:
In this project, we are trying to match restaurants from two different web sources.
Thousands of restaurant data of ten most populated US cities is crawled from
Yelp ( ~ 3000 pages)
and
Yellow Pages ( ~ 9000 pages).
Two csv files (tables) are then generated for further processing in further stages.
Using the csv files, a candidate set is obtained by blocking.
Blocking:
Blocking explanation can be found here.
Candidate set:
Candidate Tableipython code:
Blocking.ipynbMatching:
The Matching explanation is available here.
Golden Set G:
Golden Table (original)ipython code:
Matching.ipynbFinal File Submission:
Blocking
| Original | Revised |
|---|---|
| Yelp Table | Revised Yelp Table |
| YP Table | Revised YP Table |