Inspiration
In today’s digital world, fake news spreads rapidly across platforms, making it difficult for users to distinguish between real and misleading information. We wanted to build an AI-powered system that can automatically detect fake news using multimodal data.
What it does
Our project is an AI-based fake news detection system that analyzes text, images, and metadata to classify news as real or fake. It provides users with confidence scores and explanations.
How we built it
We built the system using:
Python, TensorFlow NLP models (BERT for text analysis) CNN for image verification Flask backend + React frontend Integrated APIs for real-time news data
Challenges we ran into
One major challenge was handling multimodal data efficiently. Aligning text and image predictions required building a custom fusion model. We also faced dataset imbalance issues.
Accomplishments that we're proud of
One major challenge was handling multimodal data efficiently. Aligning text and image predictions required building a custom fusion model. We also faced dataset imbalance issues.
What we learned
We learned how to build scalable ML pipelines, optimize deep learning models, and integrate multimodal data for better predictions.
What's next for AI fake news Detection
We plan to:
Improve accuracy using larger datasets Add browser extension integration Deploy as SaaS platform
Built With
- bert
- cnn
- deep-learning
- flask
- machine-learning
- natural-language-processing
- numpy
- pandas
- python
- react.js
- rest-apis
- scikit-learn
- tensorflow
Log in or sign up for Devpost to join the conversation.