Poslijediplomski studij

Zimski semestar

Prof. Nikola Bogunović

Predstavljanje znanja u informacijskim sustavima

Značenje i struktura racionalnog agenta. Paradigme rješavanja problema temeljene na agentima znanja. Hipoteza fizikalnog sustava simbola. Predstavljanje znanja i mehanizmi rasuđivanja u propozicijskoj i predikatskoj logici. Automatizirano dokazivanje teorema. Redukcija na logičke programe. Prirodno rasuđivanje u sutavima s pravilima. Postupci predstavljanja nesavršene informacije. Probabilistička logika. Sustavi s pravilima i faktorima neizvjesnosti. Bayesove probabilističke mreže i mreže za potporu u donošenju odluka. Dempster-Shafer teorija. Simboličko i nesimboličko predstavljanje znanja u strojnom učenju.

Knowledge representation in information systems

The notion and structure of a rational agent. Knowledge-based agent and problem solving paradigms. Physical symbol system hypothesis. Knowledge representation and reasoning mechanisms in logic (propositional and first-order). Automated theorem provers. Reduction to logic programming. Natural reasoning through rule-based knowledge representation and chaining inference. Current approaches to representing imperfect information. Probabilistic logic. Rule-based systems with certainty factors. Bayesian inference. Probabilistic networks (Bayesian and decision). Dempster-Shafer theory. Representing symbolic and non-symbolic knowledge in machine learning techniques.

Texts:

1. S.J.Russell and P.Norvig, Artificial Intelligence: A Modern Approach, Prentice Hall, New Jersey, 1995.

1. G.F.Luger and W.A.Stubblefield, Artificial Intelligence: Structures and Strategies for Complex Problem Solving, 3^rd ed., Addison Wesley, Reading, 1998.

2. R.J. Brachman and H.J.Levesque, Readings in Knowledge Representation, Morgan Kaufmann, San Mateo, 1985.

 Svi nastavni materijali nalaze se ovdje.