Readme_code

 -cleandata.py is for removing duplicate punctuations and emocions, such as :),!!, etc.
 -token_tag.py is for tokenizing cleaned text data with taging in NLTK.
 -getKeyword.py is for getting the keywords from tagged text data and the corresponding adjective to the keywords.
 -getkwpoint.py is for computing scores for each keywords.
 -subset_key.py is for creating a subset of keywords based on the size we provided.
 -create_loc_feature.py is for creating the location features for each business.
 -create_features_training_data.py is for creating features based on keywords. According to the previous label and location features, generate the whole feature space.
 -train_with_svm.py  Apply SVR for the features and get result.