GenAI News Verifier is a project that leverages Gemini API and News API to classify news articles as fake or real, and answer questions based on the latest and relevant news.
This project consists of two primary models:
- News Classifier: Classifies news articles into "fake" or "real".
- Question Answering Model: Provides answers to questions based on the latest and relevant news.
The project utilizes the following packages:
google.generativeai
for Gemini API integrationos
for system operationsNewsApiClient
from News API for fetching newsnltk
andspacy
for natural language processing tasks such as tokenization and stopword removal.
To set up the project:
- Register for Gemini API and News API keys at their respective websites.
- Enter your API keys in the setup section of the provided notebook (
setup.ipynb
). - Run the notebook to initialize the APIs and start using the models.
- Custom Language Model (LLM): Develop a custom language model to check facts, routing to Gemini API only when necessary.
- Full Stack Website: Create a web application to make the models accessible to all users.
- Clone the repository:
git clone https://github.com/yourusername/genai-news-verifier.git
- Install dependencies:
pip install -r requirements.txt
Detailed usage instructions and examples can be found in the notebook file (GenAI News Verifier.ipynb
).
Contributions are welcome! Please fork the repository and submit a pull request with your improvements.
For questions or feedback, please contact me.