Introduction to Machine Learning

Course
Identifier: 
COM S 474
  1. Credits and contact hours: 3 credits, 4 contact hours
  2. Instructor’s or course coordinator’s name: Chris Quinn/Forrest Bao
  3. Text book, title, author, and yearPattern Classification 2nd edition, Richard O. Duda, Peter E. Hart and David G. Stork, 2001.
  4. Other supplemental materialsMachine Learning, Tom Mitchell, 1997.

Specific course information

  1. Brief description of the content of the course: Introduction to tools and techniques of machine learning for applications. Selected machine learning techniques in practical data mining for classification, regression, and clustering, e.g., association rules, decision trees, linear models, Bayesian learning, support vector machines, artificial neural networks, instance-based learning, probabilistic graphical models, ensemble learning, and clustering algorithms. Selected applications in data mining and pattern recognition.
  2. Prerequisites or co-requisites: COM S 311, STAT 305 or STAT 330, MATH 165, ENGL 250, SP CM 212
  3. Required, elective, or selected elective? Selected Elective

Specific goals for the course

  1. Specific outcomes of instruction:
  • Appreciation of fundamental problems in machine learning
  • Ability to make intelligent choices from available algorithms subject to specific design and performance constraints, and when needed, design variants of existing algorithms. (6)
  • Ability to design, implement and evaluate intelligent agents for representative machine learning problems- e.g., learning classification rules from data, etc. (1, 2)
  • Familiarity with some current applications of machine learning
  • Ability to communicate effectively about machine learning problems, algorithms, implementations, and their experimental evaluation.

Brief list of topics to be covered

  • Review of probability theory
  • Bayesian decision theory
  • Maximum-Likelihood and Bayesian parameter estimation
  • Evaluation of classifiers
  • Naïve Bays Classifier
  • Nearest neighbor methods
  • Linear models
  • Decision trees
  • Neural networks
  • Support vector machines
  • Bayesian networks
  • Ensemble classifiers
  • Unsupervised learning: Clustering
  • Student project presentations