Yelp-Dataset-Challenge

ILS - Z534 Search Project on Yelp Dataset Challenge

Group 4 Team Members:

• Dhvani Deven Kotak (dkotak@indiana.edu)
• Manikandan Murugesan (murugesm@indiana.edu)
• Rohit Dandona (rdandona@indiana.edu)
• Vikrant Kaushal (vkaushal@indiana.edu)
• Yash Sumant Ketkar (yketkar@indiana.edu)

Task 2

Question: To help the entrepreneurs find the best location to build a successful restaurant.

Requirements:

• Python Version - 2.7.12

• MongoDB Version - 3.4

• Run pip install pymongo

• Run pip install gensim

• Run pip install nltk

• Run import nltk and nltk.download() in a python shell

• Run pip install -U jupyter

• Run jupyter nbextension enable --py --sys-prefix widgetsnbextension

• Run pip install gmaps

• Run jupyter nbextension enable --py gmaps

• python CorpusLoader.py

• Populates the reviews for Phoenix, Arizona from the dataset JSON files where type of business equals to the restaurant.

• Makes reviews more simplified for analysis by using nltk.

• python TopicModelling.py

• Gensim python library creates a LDA model for different reviews.

• python DisplayTopics.py

• Displays the six major topics and the sub-topics with maximum weightages respectively.

• All 60 topics were categorized so as to highlight the sub-topic they represent.

• The 60 subtopics highlighted in topics.txt

• python GetBusinessRating.py

• Create Ratings Collection in MongoDB.

• python SaveBusinessInfo.py

• Create Business Info Collection in MongoDB.

• python GetTop10BusinessTopic.py

• Business Frequency Topic is plotted by data generated.

• python DisplayTopicsForReview.py

• Displays the topics for review.

• Open Google_Maps_Heat_Map.ipynb in Jupyter Notebook

• Enter the topic and rating.

• Displays a heatmap of restaurants based on topic