Youtube Comment Spam Detection

This project implements various machine learning models to detect spam comments on YouTube videos.

Models Implemented

Traditional Machine Learning Models:

• Logistic Regression
• Support Vector Machine
• Decision Tree
• Random Forest
• Naive Bayes
• XGBoost

Deep Learning Models:

• CNN with BERT tokenizer
• BERT (fine-tuned)

Results

Best performing models by accuracy:

• BERT: 96.94%
• CNN: 96.68%
• SVM (Linear): 94.89%
• Logistic Regression: 94.13%
• XGBoost: 94.13%
• Decision Tree: 93.62%
• Random Forest: 92.35%
• Naive Bayes: 91.33%