Machine Learning

Course Number: LIX004M5
Instructor: Jörg Tiedemann, Alfa-informatica, tiedeman@let.rug.nl

Announcements Autumn 2007

2007-03-07: lectures moved to 9-11 on mondays, the room is "van der Leeuwzaal", Academiegebouw
2007-03-05: nobody told me that you have to activate courses on nestor. now it should be visible for everybody (I'm not going to be a big friend of nestor I have to admit ...) please, check the announcements!
2007-03-04: the course is added to nestor - I will put announcements there now! please, check that you're on the list
2007-03-03: about the labs: the situation is a bit difficult. there are no big lab rooms available at our faculty. I will try to get a Windows lab which is a bit bigger but I encourage you to bring your own laptops. you're also welcom to work from home and get the support you need via e-mail. you still need the account on hagen even when we move to the windows lab. in that case we will use ssh to connect to the machines in the unix lab to work on the assignments. furthermore, we will need to split you into 2 groups and my proposal is to have one group from 9-10:30 and the second one from 10:30-12 on fridays. hopefully we can discuss/make this split via nestor later this week.
2007-03-03: I've asked to put the course on Nestor (should hopefully be done by tomorrow), I've also asked about a larger room for the lectures on mondays - check this website to get updated information

Schedule

Sept 3 - Oct. 26
Lectures: Academiegebouw, van der Leeuwzaal, Mon. 09:15-11:00 pm
Labs: Fri. 9:15-11:00 am in Harmonie.12.107.
Exam: Friday, 26 Oct. 09:00 - 12:00 in the Academiegebouw AZERN

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 will be 3 obligatory assignments that have to be passed by the students by handing in 3 short reports according to the deadlines specified on this website. Passing these assignments is a prerequiste for the final exam at the end of the course. There will also be a final assignment which will count as 50% of the grade. This assignment is evaluated by means of a written report and an oral presentation during the last lecture of the course. The final written exam counts for the remaining 50% of the grade. 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.

Preliminary program

date	type	topic	literature/info
3 Sep	Lecture	Organization, Introduction	Ch. 1 & 2, slides
7 Sep	Lab 1	evaluation, first assignment, introduction: final projects	Ch. 5, slides, deadline first assignment: 14 Sep
10 Sep	Lecture	Decision Trees	Ch. 3, slides
14 Sep	Lab 2	select & get started with final project	mail your selection to Matthijs
17 Sep	Lecture	Instance-Based Learning	Ch. 8, slides
21 Sep	Lab 3	decision trees & IBL	second assignment, deadline: 28 Sep
24 Sep	Lecture	Bayesian Learning	Ch. 6, slides
28 Sep	Lab 4	work on final project	e-mail possible questions to Matthijs and/or me
1 Oct	Lecture	Rule Induction & Reinforcement Learning	Ch. 10 & 13 + article on RIPPER, slides
5 Oct	Lab 5	Naive Bayes & Ripper	third assignment, deadline: 12 Oct
8 Oct	Lecture	Reinforcement Learning, Genetic Algorithms & Summary	Ch. 13 + Ch. 9, slides
12 Oct	Lab	work on final assignment	deadline report: 2 Nov
15 Oct	Presentations	Student Presentations	9:00 (!) - 11:00 (van der Leeuwzaal), 11:15-13:00 (H (13)13 0309)
26 Oct	Exam		sample exam (Note: some topics such as HMM's are not relevant for this year)

Guidelines for writing lab reports

All reports must be your own work. No copying.
You may write your reports in English or Dutch. Be assured that you will not lose marks simply for making mistakes in your English (or Dutch - but I can't judge that anyway ;-))
Please submit the reports via e-mail to Matthijs
Short reports are marked as passed or not. Group work is possible with not more than 2 students per group. Passing all 3 assignments is required to sign up for the written exam.
The final assignment is marked from 1 to 10. Both, oral presentation and the written report count equally for this mark. A mark larger equal to 6 is required to pass the course.
Give concise but complete answers to the questions.
More detailed guidelines for the final assignment will be given on the website with project proposals.

About WEKA

Some information about WEKA and how to use it (old):

using WEKA
WEKAs file format
interpreting WEKA output
WEKA homepage (documentation, Wiki, ...)