Machine Learning
Course Number: LIX004M5
Instructor: John Nerbonne, Alfa-informatica, nerbonne
let.rug.nl
Off. Hours. 15:15-16 pm Mon. (after lecture)
Lab Leader: Francisco Borges, Alfa-informatica, borges
let.rug.nl
Announcements Fall 2004
- 04 Oct. 2004: The final exam is scheduled for
Friday, 29 Oct. 09:00 - 12:00 in the Academiegebouw A901
- 13 Sept. 2004: Because the room in the Harmonie building was too
small, we've moved to Oude Boteringestr. 13, room 102 effective
immediately.
- 29 July 2004: Cohen's paper on RIPPER is now
available here
Schedule
Sept 6 - Oct. 18
Lectures: Oude Boteringestr. 13, room 102, Mon. 13:15-15:00 pm. (was
Harmonie 13.342)
Labs: Fri. 9:15-11 am in Harmonie.12.102C. Four labs, starting
Fri. Sept. 24 (3rd week) (Students' Unix Room). See
Francisco's Borges's Machine Learning Labs
web site for a preview.
Exam: Friday, 29 Oct. 09:00 - 12:00 in the Academiegebouw A901
Prerequisites:
The course is open to students in Computer
Science, Artificial Intelligence and Information Science. Required
background consists of programming ability, elementary statistics,
status as 2nd year student (or higher) in study program.
Coursework, Grades, Responsibilities
There are four required
labs, which count as 20% of the grade. A written exam counts for the
remaining 80%. Students must accept the responsibilities
outlined in the general statement of
intellectual responsibility for Information Science students.
Book
Tom Mitchell Machine Learning New York:McGraw-Hill, 1997.
Overhead Sheets from the book:
www.cs.cmu.edu/~tom/mlbook-chapter-slides.html
Other Literature
-
Ian H. Witten & Eibe Frank
Data Mining: Practical Machine Learning Tools and Techniques with
Java Implementations San Francisco: Morgan Kaufmann, 1999.
We'll take the exercises from this book. The book itself is also
good, but we prefer Mitchell for its depth.
Web Site about Weka project (incl. book, software)
www.cs.waikato.ac.nz/~ml/
-
Other literature referred to in program
Exam
-
The exam is open book. Books, notes and calculators may be
used. There is a sample exam, taken
from the Weka site.
Preliminary program: (tentative)
- Organization, Inspiration Mitchell Ch.1
Intro Machine Learning Mitchell Ch.2 (Symbolic Theory
Revision, Version Spaces)
Exercises from Mitchell, Chap. 2
- Decision Trees Mitchell Ch.3
Exercises from Mitchell, Chap. 3
- Bayesian Learning Mitchell, Ch.6;
Exercise from Mitchell, Chap. 6
Lab 1 - Decision Trees
- Instance-Based Learning Mitchell, Ch.8
Example: Daelemans et al. on Dutch Diminutives
Exercise:
Lab 2 - Naive Bayes
Learning
-
Rule Induction Mitchell, Ch.10
Rules Induced from Decision Trees
Cohen's paper on RIPPER
(available here)
Lab 3 - to be announced
(IBL or RIPPER)
- Support Vector Machines Borges project.
Lab 4 -
t.b.a. (IBL, RIPPER or SVM)
- Estimation Maximization Mitchell, Ch.6; Rabin?
Example: Hidden Markov Models
Local Research on ML