MODELS OF NATURAL LANGUAGE PROCESSING FOR IMPROVING SEMANTIC SEARCH RESULTS

Abstract

Most people rely on search engines to get and share information from
all sorts of resources. All totals returned by search engines are not always important
because they are drawn from heterogeneous data sources. Moreover, it is not easy for
a user to prove that the acquired results are relevant for the request. As a result, the
semantic network plays a significant role in interpreting the relevance of search results.
This paper proposes a new method for searching for appropriate documents using
the semantic web, based on the concept of natural language processing (NLP). In the
proposed system, NLP is used to analyze a user request in terms of parts of speech. The
extracted definitions are compared with a domain dictionary to identify the appropriate
domain of the user's interest. The retrieved user request papers are examined with
natural language processing support to identify the respective domain.