All projects (assignments) undertaken as part of my Machine Learning Course (CS60050)
-
Build Decision Tree Classifer using ID3 algorithm with reduced error pruning using information gain measure.
-
Train Naive Bayes Classifier after outlier removal and calculate accuracy with 10-fold cross validation.
-
Apply laplace correction to get final accuracy.
-
Apply Principal Component Analysis for feature selection (preserve 95% of variance)
-
Apply K-means clustering over varied k values(2-8) and report value for which NMI (Normalised Mutual Information) is maximum.
-
Apply binary SVM classifier with differnet kernels to obtain best accuracy over normalised dataset.
-
Train Multilayer Perceptron models with varying density and hyperparameters. Tune the learning rate to get best accuracy.
-
Use forward sleection method for feature selection and apply ensemble learning (max-voting) to get max accuracy.