The main objective of this course is to enable the students to design and implement inductive learning systems. To this end, this introductory course gives an overview of many concepts, techniques, and algorithms in machine learning, such as inductive versus deductive reasoning, knowledge representation, classification, information gain, feature selection, supervised and unsupervised learning, overfitting and underfitting, cross validation, perceptrons, support vector machines, decision trees, nearest-neighbor algorithms, and Bayesian networks. For practical purposes the course will also get students acquainted with machine learning libraries such as Weka and Mahout.

The topics to be covered are the following: • Introduction to learning theories and machine learning • Inductive versus deductive learning • Knowledge representation • Types of inductive learning • Supervised versus unsupervised learning • Overfitting and underfitting • Cross validation • Learning with decision trees • Information gain • Learning with support vector machines • Learning with Naïve Bayes • Learning with perceptrons • Learning with kNN algorithms • Evaluating learning performances • Kappa statistics • Feature selection