This is a part of the final semester engineering project at PES University.
Given news content in the form of text, image or url, the project aims to verify the credibility of the information through the use of different metrics like:
- Fake Image Detection
- Fake Website Detection
- Comparing with Credible Sources
- Stance Detection
- Community Detection
The project aims to deliver a web application to interact with users. Along with this, the features of the project will also be exposed in the form of a REST API for developers to leverage the metrics and help fight against the spread of fake news
cd code
python3 -m virtualenv venv
source venv/bin/activate
pip3 install -r requirements.txt
python -m spacy download en
Add the heavy files mentioned below to the respective folders before trying to execute the API/web server
- glove100d.hdf5 -> code/ml/models/
- data_dump_glove.data -> code/ml/data/
To ensure PyTextrank works as expected, follow this link to make necessary changes in your installation or use the virtualenv in this repository
Run the below commands to execute both the API and website
# runs REST API
cd code/api
python3 api.py
This starts a Flask-Restful API at http://127.0.0.1:5000
# runs web server
cd code/app/client
python3 client.py
This starts a Flask-Restful API at http://127.0.0.1:9000
PS: [WIP] Formal documentation
/credible -> To compare information across multiple credible sources
- Create a class in ml/model.py which will hold all the Machine Learning model code
- Create a flask-restful class which will handle the requests in api/api/py and add appropriate endpoints for that class
- Collect input data from an HTML template using requests in app/client.py
- Link app/client.py to REST API class
- Return ML model result to new HTML file in app/client/views/ and display it there
- Akhilesh Nirna
- Aviral Joshi
- Hardik Mahipal Surana
We would like to thank our guide, Dr. S Natarajan, for his support. We would also like to thank the teachers and administration of PES University.