Machine Learning Based Domain Specific Ontology Generation Using Automatic And Authorized Learning Models

Abstract

Ontology is an important part of Semantic Web and numerous emerging AI applications. With newlinethe help of ontology, both the client and the framework can interact with each other in a newlinemachine-to-machine environment with the common understanding of the domain. Despite the newlinefact that ontology has been proposed as a vital means for representing the real world knowledge newlinefor the construction of database designs, most ontology developments are not performed newlineautomatically. However, the underlying challenges in creating and updating these Domain newlinespecific Ontologies such as need for manual intervention of Domain experts and the restrictions newlineimposed by the current technology adoptions made the tasks of automatic creation and updation newlineof Ontologies less feasible. . Therefore, the automatic generation of ontology takes very newlineimportant part in semantic web and emerging AI applications. Research Objectives newline1. To provide an algorithm to normalize and harmonize the knowledge representation across newlinedomains (IKA Intelligent Knowledge Acquisition) newline2. To design KBayes Algorithm for Ontology Learning which extract concepts, attributes, newlinevalues and relations automatically across domains (KBayes) newline3. To design an algorithm for automatically drawing relationships and / or Roles between newlineattributes (AER Automatic Entity Relationship) newline4. To design an algorithm for automatic validation of entity and relationship. (AERV newlineAutomatic Entity Relationship Validation) newline newline