Machine Learning Projects
Email Spam Classifier
The task is to determine whether and email is spam or not. I use a bag of words model to create the vocabulary and implement a Naive Bayes classifier from scratch.
See Github repo: link
Credit Card Fraud Detection
A classic class imbalance dataset - available at Kaggle.
The objective is to identify fraudulent credit card transactions. Since the class representations are highly unbalanced, confusion matrix accuracy won't do any good.
See Github repo: link
Classification on the ILPD
The Indian Liver Patient Dataset is a small dataset with slight class imbalance. The project contains some basic exploratory analysis and analyses the performance of various classification algorithms.
See Github repo: link