Extracting Sentence Elements for the Natural Language Understanding Based on Slovak National Corpus

This paper introduces an approach for extracting sentence elements from Slovak sentences based on linguistic analysis. The key idea lies in the assumption that the sentence elements relates to the meaning and they can be helpful in the process of the semantic roles identification. The system for extracting sentence elements from Slovak sentences has been developed with a morphological analyzer, disambiguator and syntactic analyzer as fundamental components. The morphological analyzer uses data obtained from the Slovak National Corpus. The syntactic analyzer uses context-free grammars. Several evaluation experiments were done with limited range of sentences for obtaining information about success of the proposed approach.