CS434: Machine Learning and Data Mining

CS 434: Machine Learning and Data Mining

Fall 2009

MWF 15:00 - 15:50 Kelly 1001

Instructor: Xiaoli Fern

Email:	xfern@eecs.oregonstate.edu
Office:	kelly 3073
Office hour:	MWF 2-3pm, or by appointment
Class email list:	cs434-f09@engr.oregonstate.edu

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.

Syllabus

Course Policy

Course materials

No text book required, lecture notes and reading materials will be posted on the webpage, please check regularly.
Useful References:

Machine learning, Tom Mitchell, McGraw-Hill 1997 (Referred to as TM).

Machine learning and pattern recognition, Chris Bishop, Springer (Referred to as Bishop).

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

see previous class for a rough lecture schedule cs434 Fall 2007; cs434 fall 2008

Date	Topics	Lecture Notes	Reading	Assignments
9/28 M	Introduction to basic concepts	slides	TM Chapter 1
9/30 W	The perceptron algorithm	slides	notes on perceptron by William Cohen	hw1; solution
10/2 F	Linear regression	Slides
10/5 M	K-nearest Neighbor, model selection	slides
10/7 W	Decision trees	slides	J. R. Quinlan, Induction of decision trees, Machine learning 1: 81-106, 1986
10/9 F	Decision trees cont	slides
10/12 M	Review of Probability Theory	slides		hw2; solution
10/14 W	Bayes classifier, Naive bayes	slides
10/16 F	Bayes classifier cont.	slides
10/19 M	Logistic regression	slides	Generative vs discriminative models	Final project Information
10/21 W	support vector machines	slides
10/23 F	SVM cont			hw3; solution
10/26 M	Ensemble learning	Slides	A short introduction to boosting
10/28 W	Ensemble learning cont.
10/30 F	Case study	Slides
11/2 M	Case study cont. Clustering.	Slides
11/4 W	Hiararchical Agglomorative Clustering			Project Proposal due
11/6 F	Kmeans	slides
11/9 M	Review		sample midterm ; solution
11/11 W	Mixture of Gaussian	slides
11/13 F	Midterm
11/16 M	Association rule mining	slides
11/18 W	Demensionality reduction	slides		hw4, cluster.csv ;rdata1; rdata2; rdata3; rdata4; rdata5;
11/20 F	Markov Decision process	slides
11/23 M	MDP cont.	slides
11/25 W	MDP cont.	slides
11/27 F	Thanksgiving holiday, no class
11/30 M	Reinforcement learning	slides
12/2 W	Reinforcement learning cont.	slides
12/4 F	TBD