Automata and Logic in IT System Modelling

5 credits

Syllabus, Bachelor's level, 1DL500

A revised version of the syllabus is available.
Code
1DL500
Education cycle
First cycle
Main field(s) of study and in-depth level
Computer Science G1F, Mathematics G1F
Grading system
Pass with distinction, Pass with credit, Pass, Fail
Finalised by
The Faculty Board of Science and Technology, 11 March 2014
Responsible department
Department of Information Technology

Entry requirements

Algebra I and Program Design and Data Structures.

Learning outcomes

To pass the course, the student should be able to

  • translate statements and reasoning given in natural language to propositional and predicate logic, explain how predicate logic formulas are interpreted as true or false, and explain the concepts of tautology, valid inference, logical truth and logical consequence
  • describe some mathematical models for calculations such as finite automata, regular and context-free grammars, and how these models are related to each other
  • use automata and logic to model simple systems and define and explain basic correctness properties of models
  • use tools to, in simple cases, determine if a model satisfies a correctness property

Content

The course consists of a theoretical part and a practical part. The theoretical part contains:

Logical expressions, formulation and specification of natural language, logical consistency and equivalence. Regular expressions, context-free and regular grammars, finite automata. System modelling. Interpretation and model. Modal logic.

The practical part contains: Modelling of a small IT system, formulation of correctness properties and analysis of systems with respect to correctness properties.

Instruction

Lectures, tutorials, labs.

Assessment

Oral and written assessment of assignments, 3 credits, written examination, 2 credits.

FOLLOW UPPSALA UNIVERSITY ON

facebook
instagram
twitter
youtube
linkedin