Kaggle link: https://www.kaggle.com/code/johnmatzakos/detectfakenewsml/notebook
Programming language: Python
Packages used:
- Pandas
- NLTK
- Scikit Learn
- Matplotlib
- and more
Results:
Classifier | Class | Precision | Recall | F1-Score | Accuracy | AUC Score |
---|---|---|---|---|---|---|
Naive Bayes | 0 | 83% | 81% | 82% | 82.49% | 89.48% |
1 | 82% | 84% | 83% | |||
Support Vector Machines | 0 | 85% | 86% | 85% | 85.70% | 85.71% |
1 | 87% | 85% | 86% | |||
Binomial Linear Regression | 0 | 85% | 85% | 85% | 92.71% | 85.59% |
1 | 86% | 86% | 86% | |||
CART Decision Tree | 0 | 86% | 82% | 84% | 84.79% | 84.67% |
1 | 84% | 88% | 86% | |||
k Nearest Neighbor | 0 | 89% | 86% | 87% | 88.15% | 88.06% |
1 | 87% | 90% | 89% | |||
Passive Aggressive | 0 | 82% | 85% | 83% | 83.77% | 83.82% |
1 | 86% | 82% | 84% | |||
Ada Boost | 0 | 87% | 88% | 87% | 87.76% | 87.75% |
1 | 88% | 88% | 88% | |||
Random Forest | 0 | 93% | 90% | 92% | 92.13% | 97.59% |
1 | 91% | 94% | 93% |