Image recognition using deep Neural Networks (NN): (Dec 2019) • Implemented and trained Logistic regression using (Python, Numpy) on ‘Cat vs non-Cat’ dataset containing 12288 features. Also, designed single layer NN and deep NN with forward propagation and backward propagation and compared their results. • Observed that deep neural network performance was best as compared to single layer neural network and logistic regression. Digit recognition using Support Vector Machines (SVM): (April 2018) • Performed classification in Python on ‘MNIST’ dataset by applying SVM with linear kernel and Radial Basis Function (RBF). • The RBF kernel gave best results with test accuracy of 97.4% whereas linear kernel test accuracy was 93.78% Classify diabetic patients using regression methods: (Feb 2018) • Executed Linear Discriminant Analysis (LDA), Quadrature Discriminant Analysis (QDA) and Ridge regression techniques on ‘diabetes’ dataset using (Python, Matplotlib) and used Mean Squared Error (MSE) as performance matrix for comparison. • From the results, concluded that the Linear and ridge regression gave lowest MSE, as the data was linearly separable.