An object--oriented approach to create a natural language understandingsystem is considered. The understanding program is a formal system built on thebase of predicative calculus. Horn's clauses are used as well--formed formulas.An inference is based on the principle of resolution. Sentences of naturallanguage are represented in the view of typical predicate set. These predicatesdescribe physical objects and processes, abstract objects, categories andsemantic relations between objects. Predicates for concrete assertions aresaved in a database. To describe the semantics of classes for physical objects,abstract concepts and processes, a knowledge base is applied. The proposedrepresentation of natural language sentences is a semantic net. Nodes of suchnet are typical predicates. This approach is perspective as, firstly, suchtypification of nodes facilitates essentially forming of processing algorithmsand object descriptions, secondly, the effectiveness of algorithms is increased(particularly for the great number of nodes), thirdly, to describe thesemantics of words, encyclopedic knowledge is used, and this permitsessentially to extend the class of solved problems.
展开▼
