https://github.com/ayushgodbole17/zomato-sentiment-analysis

A model to predict rating of a restaurant based on reviews
https://github.com/ayushgodbole17/zomato-sentiment-analysis

lstm-neural-network sentiment-analysis zomato-data-analysis

Last synced: about 2 months ago
JSON representation

A model to predict rating of a restaurant based on reviews

• Host: GitHub
• URL: https://github.com/ayushgodbole17/zomato-sentiment-analysis
• Owner: ayushgodbole17
• Created: 2022-08-25T13:33:36.000Z (over 2 years ago)
• Default Branch: master
• Last Pushed: 2022-09-06T15:17:10.000Z (over 2 years ago)
• Last Synced: 2025-02-10T01:41:44.467Z (3 months ago)
• Topics: lstm-neural-network, sentiment-analysis, zomato-data-analysis
• Language: Jupyter Notebook
• Homepage:
• Size: 169 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Zomato-Sentiment-Analysis

This project contains notebooks that help in predicting the rating for a restaurant based on its reviews  

It consists of three notebooks:  

- zomato_create_tokenizer_pickle: Since the tokenizer takes a long time to train, I've created a notebook to create a tokenizer pickle file which can be loaded when needed into the program according to a given configuration  

- zomato_transform_data: This notebook uses the original dataset and performs pre-processing to create a final dataset, which is saved as a new csv file and used for prediction  

- zomato_prediction: This notebook trains a model based on the new dataset and predicts ratings.  

Link to original dataset: https://www.kaggle.com/code/sjk714/zomato-bengaluru-analysis  

  

The pickle file, new dataset, old dataset and saved .keras model are saved in the drive folder below. You can download it and upload it to your drive to use the notebooks without change, since they were written in a colab environment.

Drive: https://drive.google.com/drive/folders/1YJHBdUKehkwT-3JBxYBhYS1cPL0LC36L?usp=sharing