This is a NLP, Deep Learning and Data Analysis Project.
- Displaying most frequently used hashtags.
- Sentiment Analysis
- Visibility Analysis i.e. calculating the reach of positive tweets vs negative tweets.
- Clone this repository.
git clone https://github.com/Shahil98/Opinion_Mining_And_Analysis.git
- Create a twitter developer account.
- Execute
pip install -r requirements.txt
to install necessary libraries. - Download dataset from
https://www.kaggle.com/kazanova/sentiment140
and place the csv file inside Classification network folder. - Execute
python generate_data.py
inside Classification network folder to generate data for training the sentiment classification network. - Execute
python network.py
to train the netwotk. After execution is completed a model.h5 and model.json file will be created. - Move to webApp folder and execute
set FLASK_APP=main.py
. - Add twitter developer credentials in main.py file.
- Execute
python main.py
to start the application athttp://localhost:5000
.