module specification

This module provides the theoretical foundations, the technologies and the corresponding tools for constructing intelligent model-driven software systems with explicit representation of knowledge. It will enable the students to model, design and implement software systems, which demonstrate “artificial intelligence” similar to the intelligence of the human behaviour. At the same time, it will help the students understand better the rationality behind the human intelligence.

The module follows one of the two main methodologies for dealing with the phenomena of Artificial Intelligence in computer science, known as “semiotic”, “symbolic” or “logical” paradigm. In this approach, the intelligent behaviour is achieved by incorporating common sense, heuristic and expert rules of behaviour which control the programmed algorithms for information processing during their execution in real time. For this purpose the module introduces a number of formal languages, such as FOL, DL and HCL, used for modelling the rational behaviour by logical methods, the corresponding mark-up languages, such as RDF, OWL and SWRL, which provide the necessary technology for representing the logical models in XML format alongside the respective software tools.

The module relies on some basic knowledge of discrete mathematics and formal logics typically taught in most of the undergraduate courses in engineering and science. Although it does not require extensive programming experience beyond the first introductory course in programing, working skills in programming using general-purpose programming languages such as Java can be greatly beneficial.

After completing this module, the students can enhance further their skills by studying the methods for automated deduction, semantic disambiguation and language translation.

N/A

• Semantic models and validation of logical theories. Satisfiability, truth and semantic equivalence. Semantic evaluation and validation.
• Inference rules and reasoning in logical theories. Entailment, logical consequence and normalization. Syntactic verification and proof.
• Using symbolic logics for specification, modelling and knowledge representation. Vocabulary of terms, facts, constraints and heuristics.
• Automated reasoning in logical theories. Unification, Resolution, Subsumption and Inheritance.
• Ontological Modelling. RDF, RDFS and OWL. Representations of individuals, concepts, attributes and properties using OWL.
• Ontological Reasoning. SWRL. Representation of regularities, constraints and heuristics using SWRL
• Querying Semantic Ontologies. Query processing using SPARQL
• Ontological Engineering. Modelling, verification and standardization of semantic ontologies. Ontology-based inference. Ontology-enabled Software Systems.
• Adding intelligence to software systems. Designing intelligent applications using model-driven software architecture
• Programmers APIs and software libraries for processing of semantic ontologies. Tools for semantic disambiguation.
• Philosophical, ethical, social and legal aspects of Ontological Engineering.

Learning Outcomes LO1 - LO4

The module combines independent study for researching the philosophical, ethical, social and legal aspects of AI with scheduled teaching activities to learn the fundamental knowledge and the practical skills needed to build intelligent ontology-enabled systems.

The blended learning is organised by combining different options:

learning resources: lecture slides, academic textbooks, public standards, workshop tasks, software tools, coursework case studies, software documentations, online discussion forums

teaching methods: lecturing, tutoring and supervising

working modes: independent, collaborative and supervised

assessment mechanisms: individual assignments, group assignments and written examinations

LO1 Obtain working knowledge of the methods for logical modelling and symbolic inference using logical languages such as Description Logic, Clausal Logic and First-order Predicate Logic
LO2 Acquire practical skills in symbolic modelling using markup languages such as RDF/RDFS, OWL and SWRL for definition of the domain terminology and formulating domain-specific heuristics
LO3 Develop expertise for designing software architectures of AI applications which utilize ontological models and implementing them using software libraries for programming ontologically-enabled systems such as Jena and Drools in general purpose software languages such as Java
LO4 Understand the challenges AI put in front of the developers from social, ethical and legal perspective

Core Textbooks:

[1] Franz Baader, Ian Horrocks, Carsten Lutz, Uli Sattler. An Introduction to Description Logic, Cambridge University Press (2017); ISBN: 9780521695428 CORE
[2] Grigoris Antoniou, Paul Groth, Frank Van Harmelen, Rinke Hoekstra. A Semantic Web Primer. MIT Press, 3rd edition (2012); ISBN: 0262018284 CORE

Additional Texts:

[3] Michael Huth, Mark Ryan. Logic in Computer Science: Modeling and Reasoning about Systems. Cambridge University Press, 2nd edition (2004);  ISBN: 052154310X ADDITIONAL
[4] Péter Szeredi, Gergely Lukácsy, Tamás Benkö. The Semantic Web Explained: The Technology and Mathematics behind Web 3.0. Cambridge University Press (2014); ISBN: 0521700361 ADDITIONAL
[5] Dean Allemang, Jim Hendler. Semantic Web for the Working Ontologist, 2nd ed. Morgan Kaufmann (2011); ISBN: 9780123859655 ADDITIONAL

Journals:

• Artificial Intelligence (Elsevier)
• Semantic Web (IOS Press)

Software:

Modelling: Protégé (https://protege.stanford.edu/)
Programming: Jena (https://jena.apache.org/)
Execution: Drools (https://github.com/protegeproject/swrlapi-drools-engine)

Software APIs:

OWL API (https://github.com/owlcs/owlapi)
SWRL API (https://github.com/protegeproject/swrlapi)

Websites:

Standards: https://www.w3.org/standards/semanticweb/
Research Articles: https://www.semanticscholar.org/
Programming Advice: https://stackoverflow.com