This repo contains work of module3 STAT 628 instructed by Professor Hyunseung Kang.
The main focus of this Yelp Data Analysis is to propose data-driven, actionable decisions to all business owner in the fast food industry to improve their ratings on Yelp. The whole project mainly consist of the following parts:
- Provide recommendations on specific business attributes
- Provide recommendations based on the reviews
- Provide personalized suggesions for all fast food resturants owners
YelpDataAnalysis_summary.ipynb contains a summary of all the aforementioned works and technical details of our findings.
The Code Folder consists the following sub-folder:
DataCleaning_Business: codes for preprocessing (extracting features and tackling missing values) the business data.
DataCleaning_Review: codes for prepocessing (extracting features and tackling missing values) the review data.
Plot_WordCloud: codes for plotting the WordCloud.
Analyse_Attribute: codes for conducting statistical analyze on business attributes.
Analyse_Review: codes for conducting statistical analyze on extracted features from the reviews.
ShinyAPP: codes for creating the Shiny APP.
The Figure Folder consists the following sub-folder:
BusinessAttributes: figures for ploting the scaled proportion of ratings with different business attributes.
WordCloud: figures of WordCloud for elementary review analysis.
MenuAnalysis: figures for ploting the scaled proportion of ratings with or without the key words extracted from the reviews.
biz_fast_food_features_all.csv: data containing all fast food resturants and their extracted features from both the business and review data.
XICI LUAN: xluan5@wisc.edu
BI QING TENG: bteng2@wisc.edu
YUHANG LAN: ylan27@wisc.edu
HONGWEI PAN: hpan55@wisc.edu