https://github.com/prajakta1321/authencheck
Amdocs Gen AI Graduate Hackathon 2024-25- A comprehensive fact-checking and misinformation detection system that leverages cutting-edge AI models and multiple news sources to verify information circulating on social media
https://github.com/prajakta1321/authencheck
api bert-fine-tuning flask-application matplotlib ngrok-server nlp nlp-machine-learning numpy pandas python3 scikit-learn seaborn wandb
Last synced: 3 months ago
JSON representation
Amdocs Gen AI Graduate Hackathon 2024-25- A comprehensive fact-checking and misinformation detection system that leverages cutting-edge AI models and multiple news sources to verify information circulating on social media
- Host: GitHub
- URL: https://github.com/prajakta1321/authencheck
- Owner: prajakta1321
- Created: 2025-02-01T11:29:21.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-02T12:28:19.000Z (8 months ago)
- Last Synced: 2025-04-08T22:38:58.520Z (6 months ago)
- Topics: api, bert-fine-tuning, flask-application, matplotlib, ngrok-server, nlp, nlp-machine-learning, numpy, pandas, python3, scikit-learn, seaborn, wandb
- Language: Jupyter Notebook
- Homepage:
- Size: 3.37 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
## AuthenCheck : Advanced Misinformation Detection System
Amdocs Gen AI Graduate Hackathon 2024-25
🎯 AuthenCheck is a comprehensive fact-checking and misinformation detection system that leverages cutting-edge AI models and multiple news sources to verify information circulating on social media. The system combines Google's Gemini AI, BERT, and traditional ML approaches with real-time news data to provide accurate fact-checking results.
✅ Features
- Real-time fact verification using Gemini AI
- Latest News verification through Google News API and BBC RSS feeds
- Historical data analysis using LIAR dataset
- Natural Language Processing (NLP) based text analysis, textblob, lemmatization, NER, context awareness, etc
- Machine Learning models for prediction
- Web interface using Flask and Ngrok
✅ Tech Stack
✔ AI/ML Models:
- Google's Gemini for advanced text analysis and NLP tasks.
- BERT for contextual understanding
- Linear Regression for prediction tasks
✔ Data Sources:
- Google News API (Real time)
- BBC News RSS Feed (Resl time)
- LIAR dataset from Kaggle (Historical data)
✔ Backend:
- Flask web framework
- Ngrok for tunneling
- Python for data processing
- Scikit-Learn
- Matplotlib, numpy and seaborn
✔ NLP Components:
- Text preprocessing
- Textblob, nltk
- Sentiment analysis
- context- awareness phrase
- Named Entity Recognition
- Topic modeling
- Tokenization
- Lemmatization
✅ Results
- Accuracy metrics
- Performance benchmarks
- Comparison with baseline models
✅ Future Improvements
- Enhanced real-time processing
- Additional news sources integration
- Mobile application development
- API endpoint creation
✅ Model Monitoring & Deployment
✔ Weights & Biases Integration
✔ Flask Web Application
✔ Ngrok Tunnel Setup
✅ Monitoring & Evaluation
- Model performance metrics tracked via W&B dashboard
- Real-time API endpoint monitoring
- System resource utilization tracking
- Error logging and alerting