This contains the personalized recommendation systems code based on Yelp data.
Author:
- Qianbo Wang
uni: qw2180 - Yi Wu
uni: yw2682 - Zuyi Wu
uni: zw2289
We download the data from Yelp Challenge Data Set: https://www.yelp.com/dataset_challenge/dataset
- Yelp_get.py contains the class of get and parse data from json file and static methods within the Yelp_get class
- Yelp_data.py contains the class of Yelp data file, with static methods of retrieving the specific features and information
- Yelp_save.py contains the methods for save and reload the data files
- Yelp_recommender.py contains the class of recommendation with methods of main mix collaborative filtering algorithm
- Yelp_main.py is the main code for building recommendation system
- Review_functions.py, Review_main, and tfidf contain the review analysis for Yelp data and get user scores from past reviews
- Yelp_Edinburgh.py contains the data processing and making recommendations for Edinburgh users
- Yelp_Pittsburgh.py contains the data processing and making recommendations for Pittsburgh users
- Yelp_Madison.py contains the data processing and making recommendations for Madison users
- First, download the data from the Yelp data set.
- Second, run Yelp_get.py to parse the data.
- Third, run Yelp_save.py to save the data into another json,pickle and csv files
- Fourth, run Yelp_Madison.py, Yelp_Pittsburgh.py and run Yelp_Edinburgh.py to get restaurants and users and also their past ratings for these specific cities.
- Finally, run Yelp_main.py, and combine the scores given by mix collaborative filtering algorithm and review analysis to calculate the final score for users and make recommendations for users.
- Then use iOS projects (objective-c) code to build the application.