note.txt

Note
--------------------------------
Business 42153
Review 1125458

Select only restaurant businesses from business object 
- 14303 businesses
- 706646 reviews
- 2141 categories
Remove restaurant with only 'Restaurants' in Categories 
- 223 restaurants
Keep only restaurants with tags other than 'Restaurants'
- 14080 restaurants
- 2140 categories
Remove 'Restaurants' tag from categories
- 2140 categories
Group restaurant by categories
Filter categories with less than 10 restaurants in group
Remove restaurant in filter-out categories
- 10595 restaurants
- 152 categories
Join with review to get text
- 469749 reviews
- 10562 restaurants
- 152 categories
Group by categories


Data Statistic
Reviews:
Restaurants: 14080
Users:
Categories: 2140

Dictionary
 - 189378 unique tokens, keeping 29212 tokens which were in no less than 5 and no more than 106 (=70.0%) documents
 - 29212 unique tokens: [u'fawn', u'mid-week', u'individual-sized', u'*best*', u'clotted']...)
 - Saved 152x29212 matrix, density=14.397% (639271/4440224)