Welcome to "True Sentiment of Food Reviews"! This project focuses on providing a more objective assessment of restaurant reviews by leveraging machine learning and sentiment analysis. The system communicates with a Python Flask environment, utilizing web scraping to fetch data from Yelp.com. The gathered information is then processed using a BERT machine learning model, and the sentiments, along with the reviews, are returned as a response to the website. The results are displayed alongside informative Highcharts bar and gauge charts.
-
Web Scraping: The Python Flask environment communicates with the Yelp.com website to fetch restaurant reviews. Due to restricted API access, web scraping is employed to acquire the necessary data.
-
BERT Machine Learning Model: The extracted reviews undergo tokenization and are processed through a BERT machine learning model. This advanced transformer model provides more objective sentiment analysis.
-
Responsive Website: The website receives user input for the restaurant name and suburb, initiates the data scraping and sentiment analysis process, and displays the results in an easy-to-understand format.
-
Highcharts Visualizations: The sentiment analysis results are presented alongside insightful Highcharts bar and gauge charts, enhancing the user experience.
-
Clone the repository:
git clone https://github.com/your-username/true-sentiment-reviews.git
-
Run the Flask application:
python app.py
-
Access the website:
Open your web browser and go to
http://{your ip}:5500/index.html.
-
Enter the restaurant name and suburb in the provided input fields on the website.
-
Click the "Submit" button to initiate the sentiment analysis process.
-
Explore the sentiment results displayed alongside Highcharts bar and gauge charts.
This project opens up exciting possibilities for future enhancements and applications, including but not limited to:
-
Auto-complete inputs and access to a broader range of reviews with increased API access or integration with different review platforms.
-
Extension of sentiment analysis to diverse applications beyond food reviews, such as assessing social media posts for real-time adjustments to engagement strategies.
-
Implementation within organizations to support multi-step approval processes for marketing strategies and proactive sentiment analysis to prevent PR blunders.
- Special thanks to the edX data analysis course for inspiring this project