# TU Wien:Knowledge-based Systems VU (Egly, Eiter, Tompits)/Oral Exam 2020-06

< TU Wien:Knowledge-based Systems VU (Egly, Eiter, Tompits)(Redirected from TU Wien:Knowledge-based Systems VU (Egly, Eiter, Tompits)/Oral Exam June 2020)

Jump to navigation
Jump to search
Egly:

- What is a Sigma-structure in FOL and what does it contain?
- What is a restricted domain; how can we restrict the domain in FOL (he wanted me to say what Herbrand interpretation for a formula is and how to construct a domain of a formula for it, in particular how to construct a domain from a formula containing no constants – essentially, we just take any one value and add it – not in the slides)?
- What is forward chaining?
- What are examples of forward and backwards chaining systems? (forward: RETE, backwards: Prolog – not in the slides)?
- Can we emulate forward chaining with backward chaining systems (yes, implementation vary but in general not too difficult – not in the slides)?

Tompits:

- How can we divide the logical programmes syntactically (describe the syntax, i.e. how the rules look like, for classical, normal, extended, and extended with disjunctions programs)?
- How to define semantics of classical LP in terms of Herbrand models and how in terms of least fixed point; what is the immediate consequence operator and what does the property of continuity mean?
- How do we start with the application of immediate consequence operator (with the empty set, then add facts, then heads of satisfied rules, etc.)
- Define syntax for normal LP; what can be said about the rules from the reduct (they are strictly positive).

There were no questions regarding probabilistic reasoning or professor Eiter’s part.

Overall, professor Egly asked questions which were not in the slides, but the grading in the end was very forgiving (at least for me, given that I had no problems with the remaining questions).