Customers write product reviews on ecommerce websites like Amazon. Amazon processes the reviews to generate commonly occurring tags. But, there exist multiple tags referring to the same thing. The Auto-generated Tags from customer Reviews are pointless and repeating like battery, battery life, battery performance etc. Reviews are not well classified as positive or negative, and same goes with tags/comments
- Comment Analyzer: Using Natural Language Processing to analyze Comments
- Tag Predictor: Predict Useful Tags Based on Comments and Classify them as Positive and Negative
- Sentiment Analysis: Using Deep Learning to Analyze the Sentiments and Mine opinions from reviews
- Chrome Extension: Products Chrome Extension that fetches reviews, whenever you visit Amazon and provides you with Tags in an interactive UI
Below is the Demo Video of our Solution located on Youtube.
- Python 🐍
- FastAPI
- Uvicorn 🌏
- Heroku
- Microsoft Azure ☁️
- HTML, CSS & JS
- Go to Release & Download latest
Tagonizer.zipfile. - Follow the below Given Illustration for setting up extension on Google Chrome running on MacOS. For other OS, it should be pretty similar.
- First Clone the repository.
$ git clone https://github.com/Ankuraxz/Tagonizer.git- Navigate into Cloned Repository.
$ cd Tagonizer- Create Virtual Environment and Activate it.
$ python -m venv venv/
$ source venv/bin/activate- Install Requirements
$ pip install -r requirements.txt- Create an Azure resource for Text Analytics. Afterwards, get the key generated for you to authenticate your requests.
- Set Environment Variable
KEY,ENDPOINT,LOCATIONwith secret token/key, endpoint/base-url and location of resource respectively. - Run the following command to start backend at
http://localhost:8000/
$ uvicorn API.main:app --reload --host=0.0.0.0 --port=8000- Open
http://localhost:8000/in browser of your choice. You will be greeted with Swagger UI and further details are present there.
This work is published under MIT License. All right reserved.
Published from India