Adaptive Systems and User Modelling course

Course at USI

Dr. Alexandra I. Cristea

(academic year 2002-2003)

Welcome to the Adaptive Systems course page for USI students! Here you can find information about the course content, schedule and assignments. This course has ended, thank you for your attendance! For any questions please e-mail a.i.cristea@tue.nl.


[ Time| Form| Evaluation| Prerequisites| Initial setting| Goals| Central principle| Schedule| Literature| Projects| Groups| MOT:Unix| MOT:download4Windows| Contact ]


Time : April 14-18 2003 (week 16); May 5-9 2003 (week 19)

 

Form:

-          Lecture (20%-25%),

-          Discussion (question hours: 5%-10%) and

-          Project work (~60%-70%)

 

Evaluation:

-          Project work evaluation (50%) - deliverable,

-          Project report (15%)- deliverable,

-          Project presentation (15%)

-          MOT evaluation (15%) – deliverable (questionnaire)

-          Course evaluation (part I: lectures; part II: project) (5%) – deliverable (questionnaire)

-          Peer evaluation– deliverable (veto[1])

 

Based on previous courses:

-          Database Technology (USI),

-          Hypermedia (USI: http://www.cwi.nl/~lynda/USI03/),

-          Mathematics,

-          Programming

 

Starting situation:

-          Number of students = 20

-          Students are at Masters level or over

-          Orientation of students: practice and experience building

-          Students’ background: various Master level studies, from computer science to philosophy

-          Expected courses studied – see above; also desirable: Mathematics at basis levels (2Y930, 2Y940, 2Y950), some probability calculus and statistics, e.g. (2S970), basic programming skills

-          Connection to students’ daily life: students study USI – so adaptive systems and user modeling from user system’s interface p.o.v.

 

Goals:

At the end of this course students should know the basic principles of adaptive systems, know at least 3 application fields for adaptive systems, should be able to design their own adaptive hypermedia system and should be able to either write the appropriately chosen content for such a system, or do the implementation of (parts of) such a system.

 

Central principle in course:

The course will discuss various aspects of adaptive systems, with a special emphasis on their many applications in information technology and education.

 

The lecture component of the course module has the following main goals:

-          Attract bright graduate students to these exciting research topics, which are well reflected in the research interests of the faculty of several UI departments, both within and outside the USI.

-          Demonstrate some of the applications of complex adaptive systems in the real world.

-          Expose students to an outstanding invited speaker of international recognition, from research and education institutions, representing some of the research areas that make up the general themes of complex adaptive systems.

The project work component of the course module has the following main goals:

-          Give the students a hands-on experience on actually building (parts of) an adaptive system.

-          Let the students develop structural thinking for design and implementation.

-          Let the students develop and familiarize themselves with collaboration strategies within a group during the group project work.

 

 

Detailed composition (program):

 

Day of the week & date

Hours

Activity

Monday 14-04-2003

9:4510:45; 11:00-12:00

13:00-14:15; 14:30-16:30

Introduction

Lecture on General Adaptive Systems

Lecture User Modeling

- Small task: data manipulation: rule construction based on UM

Tuesday 15-04-2003

9:3010:45; 11:00-12:00

13:00-15:30

Lecture on Methods:

-          Data representation + small task: concept map building

-          Data manipulation

MOT presentation

-          Theoretical frameworks: LAOS, LAG presentation

-          System demo

-          Installation discussion

Division into groups, Theme selection

Project work start

Wednesday 16-04-2003

9:3010:45; 11:00-12:45

14:00-15:00; 15:15-16:30

Subsymbolic systems talks:

-          Data manipulation for Subsymbolic Systems: NNs

-          Invited speaker: prof. Dr. Paul Cristea, Politehnica University of Bucharest: Adaptive Systems: A different perspective - Genetic algorithms

Selection of themes and sub-themes (contents discussion)

Project work

Thursday 17-04-2003

 

14:00-15:00

Project work

Groups announcing themes (contents approval)

Question hour

Handing in evaluation forms for part I (course)

Friday 18-04-2003

 

TU/e closed

Monday 5-05-2003

 

TU/e closed

 

Tuesday 6-05-2003

 

13:0014:30

Project work

Question hour (problem discussion)

 

Wednesday 7-05-2003

13:3015:30

Midway peer evaluation (oral + discussion of problem cases)

Project work

Midway demonstration of project results (+ progress discussion)

Thursday 8-05-2003

 

13:3014:30

Project work

Question hour

Handing in evaluation forms part II (project work)

Handing in evaluation forms for MOT (part III)

Friday 9-05-2003

9:00-10:00

 

 

 

 

13:30-17:00

Handing in reports (MOT-analysis & program files: per e-mail)

Handing in final peer evaluation (only veto’s: e-mail)

Demonstration of project results

Handing in project results (floppy or e-mail)

Presentation of projects

??–05-2003

(to be announced to secretary)

 

Notification of grades

 

 

Literature:

-          LECTURES:

         Powerpoint presentation files of lectures are available on-line.

-          THE SYSTEM FOR THE PROJECT WORK:

         MOT (MOT 3.0) is available for download for Windows,

         and as well as a running system on Unix.

         MOT 3.1 is currently available!! It has a more compact format, but also a testing function for database connectivity between Perl and MySQL.

-          QUESTIONNAIRES:

         Questionnaire Part I: Course (if not yet delivered: deliverable for Thursday)

         Questionnaire Part II: Project (NEW!! deliverable for Thursday)

         Questionnaire Part III: MOT (NEW!! deliverable for Thursday)

-          ANALYSIS INFO:

         MOT-analysis (NEW!! deliverable for Friday)

 

Non-obligatory literature:

General Adaptive Systems:

-          Peter Brusilovsky, Adaptive Educational Systems on the World-Wide-Web: A Review of Available Technologies: http://manic.cs.umass.edu/~stern/webits/itsworkshop/brusilovsky.html

-          IOWA College of Business, Complex Adaptive Systems and their Business Applications: http://www.biz.uiowa.edu/class/6K299_menczer/

-          Links to material on Intelligent Interfaces: http://www.sics.se/~annika/ii_links.html

-          SSIE 580B: Evolutionary Systems and Artificial Life: http://www.c3.lanl.gov/~rocha/ss504_02.html

-          Adaptive Hypertext and Hypermedia homepage: http://wwwis.win.tue.nl/ah/

-          Active Web’99 Proceedings on-line: http://www.visualize.uk.com/conf/activeweb/proceed/contents.asp

-          Evolving artificial creatures, Karl Sims: http://biota.org/ksims/blockies/index.html#video

-          Example genetic algorithms: http://www.cs.bgu.ac.il/~sipper/biomorphs/evolution.html

 

Agents:

-          Steve: http://www.isi.edu/isd/VET/steve-demo.html

-          Intelligent Agents Repository: http://www-cia.mty.itesm.mx/~lgarrido/Repositories/IA/index.html

-          Is it an Agent, or just a Program?: A Taxonomy for Autonomous Agents: Stan Franklin and Art Graesser, Institute for Intelligent Systems,University of Memphis: http://www.msci.memphis.edu/~franklin/AgentProg.html

 

User Modeling:

-          Alfred Kobsa (1990: old paper), User Modeling in Dialog Systems: Potentials and Hazards, SFB 314: AI – Knowledge-Based Systems, http://www.ics.uci.edu/~kobsa/papers/1990-AISoc-kobsa.pdf

-          User Modeling journal (new papers): http://umuai.informatik.uni-essen.de/

-          User Modeling homepage: http://www.um.org/

-          Emotional modeling: http://www.ict.usc.edu/~gratch/

-          Emotionware: http://www.acm.org/crossroads/xrds3-1/emotware.html

-          UM’03 homepage: http://www2.sis.pitt.edu/~um2003/

 

Neural Networks:

1. An Introduction to Neural Networks, Kevin Gurney, UCL Press, 1997.

2. Artificial Neural Networks, Dan W. Patterson, Prentice Hall, 1996.

3. Neural Networks (Second Edition), Simon Haykin, Prentice Hall, 1999.

4. Fundamentals of Artificial Neural Networks, Mohammed H. Hassoun, MIT Press, 1995.

5. Introduction to the Theory of Neural Computation, John Hertz, Anders Krogh, Richard G. Palmer, Addison-Wesley, 1991.

1)       and 2) are introductory books, while 3), 4) and 5) are more extended and treat in-depth mathematical aspects.

 

Other NN related links:

-          NN, GA, AI: http://www.brunel.ac.uk/~csstgdm/lecture1.pdf

 

UML tutorials:

-          http://www.togethersoft.com/services/practical_guides/umlonlinecourse/#statechart-diagrams

-          http://pigseye.kennesaw.edu/~dbraun/csis4650/A&D/UML_tutorial/index.htm

 

Concept Maps:

-          http://www.udel.edu/chem/white/teaching/ConceptMap.html

 

Papers on LAOS, LAG, MOT:

-          http://wwwis.win.tue.nl/~alex/HTML/Minerva/papers/

 

Demos:

-          Neural Networks

-          Ants (download)

 

Other Links:

-          Moshe Sipper: http://www.moshesipper.com/caslinks.html

 

 

Project themes:

(delivery of project results via email to a.i.cristea@tue.nl ; database files should be dumped (with the mysqldump function) into files with significant names then sent as attachment; description files should be Word documents, HTML or PDF. Please note that the project delivery only represents a part of the final grade. Check ‘Evaluation’ for the grade composition.)

 


  1. Writing a course/presentation on Adaptive Systems (or subject to be discussed with me) with MOT (maximal grade: 9-10; maximal group size: 2-3):
    1. Cancelled because of existing on-line version (running system on Unix): Installing MOT (+ Apache, Perl, Javascript)
    2. Trying out existing course manipulation with MOT
    3. On paper: Analyzing MOT from the p.o.v. of data representation and manipulation principles (for adaptation) (min page - max 1page): MOT-analysis (table 1: NEW!!)
    4. On paper: Analyzing MOT from USI perspective (definition of desired USI proprieties with respect to the goals & analysis of completion & suggestions for completion) (min page - max 1page): MOT-analysis (table 2 NEW!!)
    5. In MOT system: Designing the new material according to the MOT principles
    6. In MOT system: Introducing the new material in MOT according to these principles (DM + GM)
    7. On paper: Writing a short description of the UM that should be added and its variables (min 2 – max 10 user model variables descriptions:
      • <variable-name; domain; global/local>; e.g. <knowledge; 0-100;local>; or <emotion; happy-sad-indifferent;global>)
    8. On paper: Writing a short description of the AM using the above, as well as illustrative examples with UML state diagrams or other diagrams or rule descriptions that show the systems dynamics expected (rules as done in the exercises of the first week; min 10 – max 20 rules, out of which min 2 generic and min 2 specific rules)
      • Ex. Specific: IF UM.neural-network.text.access=TRUE AND UM.neural-network.knowledge=0 THEN UM.neural-network.knowledge=100;
      • Ex. Generic: IF concept.access=TRUE AND concept.knowledge=0 THEN concept.knowledge=100;

 

 


  1. Writing a new interface module for MOT (maximal grade: 10; maximal group size: 3):
    1. Installing MOT (+ Apache, Perl, Javascript): please install newer version: MOT 3.1; more info at: here
    2. Trying out existing course manipulation with MOT
    3. On paper (optional): Analyzing MOT from the p.o.v. of data representation and manipulation principles (for adaptation): MOT-analysis (table 1: NEW!!); MOT-analysis (table 2 NEW!!)
    4. Designing the module according to the MOT principles: the interface should represent other (interesting) views on lessons (or concepts), e.g.:
      • Treat AND/OR connections at lesson level differently
      • Show also non-standard attributes
      • Interpret HTML tags
      • Interpret existent formatting (<CR> and spacing)
      • etc.
    5. In MOT system: Introducing the new module in MOT (i.e., programming it)
    6. Testing the new module

 


  1. Writing a course on Adaptive Systems (or subject to be discussed with me) with MOT and the new interface module to display its facilities (maximal grade: 10; maximal group size: 5-6):
    1. Installing MOT (+ Apache, Perl, Javascript)
    2. Trying out existing course manipulation with MOT
    3. Analyzing MOT from the p.o.v. of data representation and manipulation principles (for adaptation)
    4. Designing the new material according to the MOT principles
    5. Designing the module according to the MOT principles and the principles of the new material
    6. Introducing the new module in MOT
    7. Introducing the new material in MOT according to these principles
    8. Testing the new module with the new material

 

Other project information:

Project groups (please note that the size of the group also influences the expected results):

1.      Pavel, Olivier, Anita

a.       Project selected: (1)

b.      Project theme: “Concept Maps for Adaptive Systems”

2.      Ivo, Harold, Hans

a.       Project selected: (1)

b.      Project theme: “Intelligent Tutoring Systems”

3.      Hanneke, Wouter, Wen

a.       Project selected: (1)

b.      Project theme: “Overview of Adaptive Hypermedia”

4.      Claudia, Hienadz, Judith, Privender

a.       Project selected: (1)

b.      Project theme: “E-learning”

5.      Javed, Andres

a.       Project selected: (2)

b.      Project theme:

6.      Siska, Lyubov

c.       Project selected: (1)

d.      Project theme: “Introduction to Artificial Intelligence”

 

 

Questions, etc., to:

a.i.cristea@tue.nl

 

Further contact details:

Dr. Alexandra Cristea, assistant professor, Databases and Hypermedia group, HG 7.82

Information Systems Department, Faculty of Computer Science and Mathematics

Eindhoven University of Technology, Postbus 513, 5600 MB Eindhoven, The Netherlands

Ph.: +31-40-247 4350 (direct)

Ph.: +31-40-247 2733 (secretary)

Fax: +31-40-246 3992

http://wwwis.win.tue.nl/~alex/

 



[1] Each student should also evaluate the other students’ in her/his group and give a passing/not passing grade. This is to avoid that some students don’t work at all.