Introduction to Machine Learning
Understand the learning system model, training, testing, performance, Machine learning structure and the various learning techniques.
Nearest Neighbor Classification
Know about the instance based classifiers, nearest-neighbor classifiers. Master the difference between lazy and eager learning, understand k-NN variations, learn how to determine the good value for k and when to consider nearest neighbors. Learn condensing nearest neighbor issues and nearest Bayes classification.
Naive Bayes Learning
Learn conditional probability, master the basics of the Bayesian theorem, Bayes classifier, model parameters, naive Bayes training, types of errors, sensitivity and specificity, ROC curve, holdout estimation and cross-validation.
Understand key requirements, decision tree as a rule set, how to create a decision tree and choosing attributes, ID3 heuristic, entropy, tree induction, splitting based on ordinal attributes. Determine the best split and the strength and weakness of decision trees.
Learn binary response regression model, linear regression output of proposed model and work on the problems with linear probability model. Understand logistic function, logistic regression, its interpretation, odds ratio, goodness of fit measures, confusion matrix.
Introduction to Cluster Analysis
Gain insight of types of data in cluster analysis, categorization of major clustering methods, partitioning methods, hierarchical methods, density-based methods, grid-based methods, model-based clustering methods and supervised classification.
Principal Component Analysis (PCA) and Forecasting Principles
Realize the curse of dimensionality, dimension reduction. Understand the importance of factor and component analysis, principal component analysis, basic time series and its components. Learn about the moving averages (simple & exponential), R’Â’s inbuilt function ts(), plotting of time series, business forecasting using moving average methods, the ARIMA model and the various application of ARIMA model in business.