IBM ML Project

---

Team Members

• Nithin Krishna

---

Machine Learning Model Used

I have used KNN , Decision Tree Classifier , Random Forest & XG Boost algorithms on the KDD dataset which when compared to the deep learning models has given us quite low accuracies.

SMOTE

I have applied the SMOTE algorithm in order to generate new duplicate samples. I have performed a case where I've upsampled the minority class in order to balance the dataset.

Deep Learning Model Used

I have also implemented one of the deep learning models i.e ANN which has given me the highest accuracy which is around 0.99.

Association Rule Mining Model Used

Built an Apriori which is an algorithm used for Association Rule Mining. It searches for a series of frequent sets of items in the datasets.