Fake News Detection using ML & CNN
Mar 1, 2025
·
1 min read

A hybrid fake news detection system combining traditional machine learning with deep learning techniques. The project focuses on identifying misinformation by using both TF-IDF + Naive Bayes pipelines and Convolutional Neural Networks for text classification.
Key Features:
- 🧹 Debiased the dataset to reduce skew and ensure fair model performance
- 🔍 Built a classic ML pipeline using
TfidfVectorizer
andMultinomialNB
for efficient fake news classification - 🧠 Implemented a CNN-based NLP model for deeper context understanding and semantic pattern learning
- 🔄 Compared performance between traditional and deep learning methods to ensure robustness
- 📊 Visualized metrics like accuracy, precision, recall, and confusion matrix for model evaluation
Technologies Used:
- Scikit-learn (for pipeline, TF-IDF, Naive Bayes)
- TensorFlow / PyTorch (for CNN)
- Pandas, NumPy, Matplotlib, Seaborn (for data preprocessing & visualization)