This project idea is taken from Cousera Andrew Ng course of Machine Learning
#Exercise 1 Basic matlab functions are implemented. Linear regression with one variable is implemented. Data set of population of a city is given along with profit associated with it. Prediction is done using linear regression which city should be chosen next to maximise the profit. Parameters are made to fit using Gradient Descent algorithm.To check whether gradient Descent is working properly the graph of Cost function is plotted along with the contour applying gradient descent at every step.
In ex1_multi linear regression using multiple variables is implemented.Features are normalised before implemeting algorithm. The same prob is solved using Normalization to check the answers.
#Exercise 2 Logistic Regression is implemented to predict whether a student can be admitted to a particular university by using the two scores.
#Exercise 3 Digit Recogntion first implemented by logistic regression. Then it is implemented by neural network. The feed forward part is in this exercise.
#Exercise 4 Backpropagation is implemented to identify digits using neural network.
#Exercise 5 This exercise is about bias-variance problem which can be seen by plotting learning curves. First part of the exercise ,regularized linear regression is implemented to predict the amount of water flowing out of a dam using the change of water level in a reservoir.
#Exercise 6 SVM is implemented in a dataset and various plots are seen with changing parameter C.In the first part SVM is implemented with gaussian kernel In the next part of exercise Spam Classification is done using SVM.The dataset included for this exercise is based on a a subset of the SpamAssassin Public Corpus
#Exercise 7 K-means Clustering is done to implement image compression.In the next part dimensionality reduction is done using PCA on a face dataset.
#Exercise 8 In the first part of exercise anamoly detection is done using gaussian model.In the next part collaborative filtering algorithm is implemented for recommender system of movies.