Repository containing portfolio of data science projects completed by me for academic and self learning purposes.
Presented in the form of Jupyter Notebooks.
Note: Data used in the projects is for demonstration purposes only.
-
Supervised Machine Learning
- Machine Learning with Linear Regression : Using Linear Regression to predict decline rate of honey bee.
- Machine Learning with Logistic Regression : Using Logistic Regression model to predict which passengers survived the sinking of the Titanic. Uses publically available data from Kaggle.com.
- Machine Learning with Multiple Linear Regression : Using Multiple Linear Regression to predict the approximate rent of the apartment using variables that factor in to pricing a home.
- ML with K Nearest Neighbours : Using K Nearest Neighbour to predict whether a patient has breast cancer.
- Machine Learning with Support Vector Machine : Using Support Vector Machine to find the decision boundary of the strike zone in baseball. (Sports Vector Machine)
- Machine Learning with Decision Trees : Using Decision Trees to find the best features which enable to predict the continent from which a given flag comes from.
- Machine Learning with Random Forests : Using Random Forests to predict whether a person makes more than a threshold income.
-
- Gradient Descent and Linear Regression : A basic project which showcases implementation of partial derivative and of Linear Regression using Object Oriented Programming approach.
- Perceptron : Implements the most basic form of a neural network, a perceptron and uses a linear model to classify 2 regions of data.
- Deep Neural Networks : Uses a non-linear model to find boundary between the 2 datasets and trains a deep neural network.
- Image Recognition - MNIST : Trains a deep neural network on MNIST image dataset to identify the handwritten digits in a batch of 20.