CT481 Logics and Databases
Rationale
In most database design methodologies - and in particular when designing
these new generation systems - the conceptual modelling of the application
domain plays an important role. Good conceptual data models put their
emphasis on the correct and semantically rich representation of complex
relations that may exist between data. A logical characterisation of
conceptual data models helps in understanding the foundations and in
mastering the design methodologies for complex information systems.
Aims
The aim of this module is to give the students an understanding of
conceptual data modelling in databases in terms of formal logic. The
course is divided in three major parts. The first part of the course will
review the basic concepts of classical first order predicate logic, and
will give the student the ability to model reality using classical logic.
The second part introduces the Extended Entity Relationship (EER) and UML
Class Diagrams standard conceptual data models, reviewed from a logical
perspective. The third part of the course will present the most popular
logic-based conceptual modelling formalism, namely Description Logics. The
simplest Description Logic will be deeply analysed from the logical point
of view. Several extensions and uses of Description Logics will be briefly
introduced at the end. The relationship between EER/UML and Description
Logics will be covered. Illustrations of practical examples will be given.
Learning Outcomes
Students will be able to:
- model reality using classical logic, and understand how deduction can
be performed automatically;
- understand the logical foundations of the Extended Entity Relationship
(EER) and UML Class Diagrams
- master the most popular logic-based conceptual modelling formalism,
namely Description Logics;
- build knowledge bases using Description Logics;
- understand the relationship between EER/UML and Description Logics.
Reading List
- (A) "The essence of logic", John Kelly. Prentice Hall, 1997.
- (B) "Logic", Schaum's Outlines by J. Nolt, D. Rohatyn, and A. Varzi,
McGraw-Hill, 1998.
- Various scientific articles on the topic will be provided during the
course.
Dependencies
IMPORTANT: Pre-requisites for this module:
- some knowledge of logic (preferable);
- GOOD
knowledge of relational databases and Entity Relationship modelling (compulsory);
- some knowledge of UML modelling (preferable).
Assessment
Examination: 70%
Course work: 30%
Example Material from last year course (CS3411)
Check the slides from the 1999
CS3411 course material, for the parts related to propositional and
first-order logic.
Course change option
Students have the option to take CT314
- Emerging Technologies in Information Management instead of this
course.
Make your choice by the 20th of November, 9:00AM.
Lectures and course slides
Room: MB/C-53, 9-12am.
- Monday, 25 September, 9:00am
Introductory
lecture
- Monday, 2 October, 9:00am
Foundations
of Propositional Logic
- Monday, 9 October, 9:00am
Reasoning with Propositional Logic
- Monday, 16 October, 9:00am
Foundations
of First-order Logic
- Monday, 23 October, 9:00am
Foundations
of First-order Logic
- Monday, 30 October, 9:00am
No lecture (sorry, I was ill!).
- Monday, 6 November, 9:00am
Using First-order Logic
- Monday, 13 November, 9:00am
Using
First-order Logic and (Re-)Introducing
ER and UML modelling.
- Monday, 20 November, 9:00am
Assessed individual class exercise.
- Monday, 27 November, 9:00am
The
Meaning of Class Diagrams and ER Schemas
- Monday, 4 December, 9:00am
Correction of the class exercise, and First
Order Logic and ER Schemas
- Monday, 11 December, 9:00am
Revision.
Useful Resources and Links
Enrico
Franconi, University of Manchester, Department of Computer
Science,
franconi@cs.man.ac.uk
Last modified: Mon Dec 11 11:52:48 GMT 2000