KRR
Knowledge Representation and Reasoning
COMP5450M
Handbook Page,
Catalogue
Page
The lecturer for this course is
Brandon
Bennett of the School
of Computing.
Objectives:
On completion of this module, students should be able to:
analyse informal descriptions of problems and
real world situations in terms of a number of different formal
representation languages;
use an automated reasoning software tool to compute inferences from
logical representations;
understand the basic principles of automated reasoning
and appreciate its power and its limitations.
Topics Covered:
- Classical Logic and Inference
- Revision of propositional and 1st-order logic
- A sequent calculus proof system
- Basic model theory
- Soundness, Completeness and Decidability
-
Representing and reasoning about time and change
- Tense Logic and 1st-order temporal logic
- AI representaions of time
- Non-monotonic reasoning about change
-
Space and physical objects
- Qualitative spatial relationships
- Matter, objects and physical events
- Combining space with time
-
Ontology and AI Knowledge Bases
- The need for world knowledge
- Ontology and Ontologies
- Description Logic
- The problem of vagueness
Lectures and Additional Course Material:
The lectures for this course are available here (in PDF format).
The lectures from 2007 can be found here.
Supplementary course material:
-
A Concise Introduction to the Fundamentals of Classical Logic
(pdf,
2-up pdf),
.
-
An excersise on translating from English to 1st-order logic
(pdf).
-
An example using Sequent Calculus to show the inconsistency inherent
in Russell's famous set-theoretic paradox
(pdf).
-
Supplementary (borrowed) lecture notes and slides:
-
Learn Prolog Now!
Online notes and tutorial on the Prolog programming language.
- Knowledge Representation and Reasoning, Leora Morgenstern
(course web site)
- Introduction to the Situation Calculus, Maurice Pagnucco
(PDF slides)
- Introduction to Ontologies and Description Logics, Carlos Areces
(PDF slides)
- From Tableaux to Automata for Description Logics
(PDF slides)
[Introductory material on slides 1-7 is most relevant]
Example exam papers:
- 2002 paper
(PDF)
- 2004 paper
(PDF)
- 2005 paper
(PDF)
- 2006 paper
(PDF)
- 2007 paper
(PDF)
-
Past paper answers
I strongly recommend that you make a concerted attempt
to answer the questions before looking at the answers,
otherwise you will not learn the thought process required
to solve the problems on your own. Note that for some questions
there may be alternative, equally correct answers.
Online Discussion Groups:
An excellent way to get help on any topic relating to this
module is through the on-line discussion groups on the Computer Studies
server (lnntp.scs.leeds.ac.uk). You can post queries, criticism, or you
can just have a general discussion. Using your news reader (e.g. Netscape,
News Express, gnus) you should subscribe to the following two discussion
groups:
local.modules.krr
and local.modules.krr.talk
The first of these is for important announcements about
the module.
The second is for you to post any kind of comment or
query relating to the module.
Coursework:
-
Coursework 1 Automated Reasoning --
20% of total marks for course.
Set 21st October, Due in 9am, 14th November, 2008.
To complete this you will need to use both the Prolog implemented
Gentzen theorem prover and the Otter theorem prover.
See the Software Resources section below for further
details about these.
-
Coursework 2 (2007)
Build your own Ontology -- 20% of total marks for course.
Set 18th November, Due 9am, 9th December, 2008.
A sketch of an ontology specification for the domain of Sailing
is given here.
Software Resources
- A Prolog Gentzen Theorem Prover
A Gentzen calculus theorem prover written in
Prolog can be found at ~krr/gentzen_prover_2008.pl.
An exercise explaining how to use it is
available here.
- The Otter Theorem Prover
An executable of the Otter theorem prover is available
at ~krr/otter.
An exercise explaining
the basics of using Otter is available
here.
Reading Material:
There is no set text for the module. The following books
may be useful:
-
Computational Intelligence: A Logical Approach, David Poole
Alan Mackworth Randy Goebel, OxfordUniversity Press, New York, 1998.
-
Russell S and Norvig P (1995), Artificial Intelligence: A
Modern Approach, Prentice Hall.
-
E Rich and K Knight (1991), Artificial Intelligence (2nd
Ed), McGraw Hill.
-
M Ginsberg (1994), Essentials of Artificial Intelligence,
Morgan Kaufmann.
-
T Dean, J Allen and Y Aloimonos (1995), Artificial Intelligence:
Theory and Practice, Benjamin Cummins.
-
Readings in Knowledge Representation, Editors: R J Brachman
and H J Levesque, Morgan Kaufmann, 1985.
-
Representations of Commonsense Knowledge, E Davis, Morgan
Kaufmann, 1990.
From time to time I will make recommendations for specific
additional reading from research papers or book chapters.
Some useful links: