Study on implementing search techniques on lexical resources

Abstract

The insight of the Semantic Web introduced ontologies as the main integrating tool for the management of the knowledge and semantic structure of various documents. This research work describes the hurdle to combining freely available online materials to build inclusive lexical resources. It also provides a useful user interface and RESTful API to access and manipulate the lexical resource using the proposed methodology. The design and implementation of the LKE (LearnITy Knowledge Engine) system capability modules focus on recognizing and analyzing the most suitable and easy approach for extracting and processing different multilingual WordNet (also used as ontology) and ontology files. To extract relevant information such as synonyms, hyponyms, hypernyms, classes, and properties, from these files, we can utilize various API's such as MIT Java Wordnet Interface (JWI), WOLF, and Web Ontology Language (OWL). This extracted knowledge can then be linked and utilized for testing purposes in the search interface. The data obtained after extraction is stored in separate tables, creating its own LKEApplication database. This database is used to store the extracted files and data from WordNet and Ontologies. This research involves several steps, which includes uploading various wordnet and ontology files (such as PWN, DebVisDic, XML-based, .owl, .rdf), extracting data, parsing the data using different APIs, and analyzing the results. This analysis is highly beneficial for text and word analysis, aiding in the development of the proposed ontology search tool. It can be efficiently exploitable in different artificial intelligence applications. Besides all the mentioned features, this ontology extraction algorithm is created keeping in mind relatable possible usage scenarios of the ontologies for understanding and program analysis