Description: robot-brain.JPG



CS 434: Machine Learning and Data Mining

 Fall  2012

MWF 12:00 - 12:50 Kelly 1003




Instructor: Dr. Xiaoli Fern



kelly 3073

Office hour:

MWF 11-12pm, or by appointment

Class email list:


Machine learning and Data mining is a subfield of artificial intelligence that develops computer programs that can learn from past experience and find useful patterns in data.  This field has provided many tools that are widely used and making significant impacts in both industrial and research settings. Some of the application domains include personalized spam filters, HIV vaccine design, handwritten digit recognition, face recognition, credit card fraud detection, unmanned vehicle control, medical diagnosis, intelligent web search, etc.

This course will provide a basic introduction to this dynamic and fast advancing field. Topics include the three basic branches in this field: (1) Supervised learning for prediction problems (learn to predict); (2) Unsupervised learning for clustering data and discovering interesting patterns from data (learn to understand); and (3) Reinforcement learning for learning to select actions based on positive and negative feedback (learn to act). It will have a special focus on the practical side --- students will not only learn various machine learning and data mining techniques, but also learn how to apply them to real problems in practice.


Course Policy

Course materials


Students are expected to have the following background:

·        Knowledge of basic computer science principles and skills.

·        Familiarity with the basic probability theory (Stat314 should be sufficient but not necessary).

o   Slides for review of basic probability concepts

·        Familiarity with vector calculus (MTH254 should be sufficient but not necessary) and linear algebra

o   Slides for review of basic concepts in vector calculus and linear algebra


Learning objectives

Upon completing the course, students are expected to:

1) be able to apply supervised learning algorithms to prediction problems and evaluate the results.

2) be able to apply unsupervised learning algorithms to data analysis problems and evaluate results.

3) be able to apply reinforcement learning algorithms to control problem and evaluate results.

4) be able to take a description of a new problem and decide what kind of problem (supervised, unsupervised, or reinforcement) it is.


Lecture Schedule

Week 1 (9/24/12-9/28/12) (Reading assignment: please read the review slides posted under Prerequisites above)

·        Lecture 1 Introduction to basic concepts slides

·        Lecture 2 Linear regression slides

·        Lecture 3 Linear classifier – perceptron slides
Assignment 1 (Please email the instructor your answer to the first written question directly. Use CS434-hw1 in your title of the email.)Data for assignment 1: training; testing . Note that the data is tab delimited. The first column stores the class labels, which are either 1, or -1. Additional data for further testing your perceptron algorithm.
Solution to assignment 1

Week 2 (10/1/12-10/5/12)

·        Lecture 4 Lazy learning: k-nearest neighbor slides

·        Lecture 5 & 6 Decision tree  slides

Week 3 (10/8/12-10/12/12)

·        Lecture 7 Naive Bayes Classifier slides

·    Assignment 2 Data for assignment 1: training; testing . Note that the data is comma delimited. The first column stores the class labels, which are either 1, or 0.

Week 4 (10/15/12-10/19/12)

·        Lecture 9 Naive Bayes Classifier cont.

·        Lecture 10 Logistic regression: slides

·        Lecture 11 SVM: slides

Week 5 (10/23/12-10/27/12)

·        Lecture 12 Unsupervised learning, clustering: Kmeans slides

·        Lecture 13 Clustering: HAC slides

·        Lecture 14 Clustering: GMM slides

Week 6 (10/29/12-11/2/12)

·        Lecture 15 Association rules slides

Week 7 (11/5/12-11/9/12)

·        Lecture 16: guest lecture. The Impact of Visual Appearance on User Response in Online Display Advertising

·        Lecture 17: Dimension reduction: PCA Slides

·        Lecture 18: Ensemble learning Slides

Week 8 (11/12/12-11/16/12)

·        Lecture 19: RL: Markov Decision Processes Slides

·        Lecture 20: Markov Decision Processes - value iteration and policy iteration Slides

Week 9 (11/19/12-11/23/12)

·        Lecture 20: Reinforcement learning: passive learning Slides

Week 10 (11/26/11-11/31/12)

·        Lecture 21: Reinforcement learning cont. Slides

·        Lecture 22: Case study

·        Lecture 23: Overview of advanced ML topics