Dynamic Modelling for Human-Centered Systems

2019-2020

Course Objective

At the end of the course, the student has knowledge and understanding
of: the role that dynamic models can play in human-centered systems; the
steps required to develop a dynamic model; different representations and
implementations that can be used for dynamic models.

The student is able to apply this knowledge and understanding to: build
dynamic models based on a textual description of a process in a domain
(e.g. psychology); integrate a dynamic model in an intelligent system to
derive conclusions about the state of the environment and/or user;
integrate a dynamic model in an intelligent system to decide on relevant
actions.

The student is able to making judgments about: the appropriate
representation for a dynamic model; the validity of a dynamic model
based on simulation results.

The student has acquired communication skills to: report in a scientific
and precise manner about the design and the evaluation of a dynamic
model.

The student has acquired learning skills to: read and interpret
semi-scientific texts from other domains (e.g. psychology, sociology).

Course Content

Intelligent systems usually encode specific information about the
context in which the system executes, e.g. the users, the physical
environment and the task environment in which it is used. This
information does not only cover environmental states, but also
information on the various processes in the environment. Dynamic models
are a useful way to encode these processes.

In this intense 8-week bachelor course, the students will learn how to
develop dynamic models based on literature about process related to
human functioning. The course considers methodological aspects of
modelling, such as: the collection and specification of relevant
knowledge, the definition of relevant ontologies, the definition of
simulation experiments and the validation of the results obtained from
the modelling process. The course also considers fundamental aspects of
dynamical modelling methods, including: defining causal relationships
and using causal graphs, defining/using executable specifications,
differentiating and integrating quantitative and qualitative models.


In addition, a methodology will be taught on how to use such dynamic
models as basis for intelligent, human-centered systems. Based on
realistic examples from psychology (where emotions or moods are
modelled), bio-medicine (where physiological models of the body are used
to measure intoxication), or sociology, it is shown how dynamic models
of such a domain can be incorporated to derive conclusions about the
current situation (assessment) or to decide on relevant actions
(support).

During the course, student will combine the completion of weekly
assignments based on lectures with the results of constructing models
that study self-selected problems. The latter models will be validated
via software systems that support analysis and simulation. The student
will be challenged to relate the models to relevant external input
information (including, where appropriate, sensor-based inputs).

Teaching Methods

Lecture and practical sessions.

Method of Assessment

Practical assignments (50% of the grade) and exam (50% of the grade).

No resit is possible for the practical assignments.

Target Audience

B Artificial Intelligence (year 1)

General Information

Course Code XB_0036
Credits 6 EC
Period P4
Course Level 100
Language of Tuition English
Faculty Faculty of Science
Course Coordinator dr. M.C.A. Klein
Examiner dr. M.C.A. Klein
Teaching Staff dr. M.C.A. Klein

Practical Information

You need to register for this course yourself

Last-minute registration is available for this course.

Teaching Methods Lecture, Computer lab
Target audiences

This course is also available as: