The project contains classification of spam and non-spam messages of the Malayalam Language. Various ML Models have been used for the classification like Logistic Regression, Support Vector Machine, Decision Trees, Multinomial Naive Bayes, Random Forests and many more. Among which Logistic Regression and Random Forests gave best results.
The following are the results:
| Model | Accuracy | Micro average | Weighted Average |
|---|---|---|---|
| Random Forest | 0.74 | 0.66 | 0.76 |
| LogisticRegression | 0.73 | 0.72 | 0.72 |
The results have been taken considering f1 score.