Semi-Automated Ontology Building Using Deep Learning to Provide Domain-Specific Knowledge Search in the Marathi Language

Abstract

In this work, our goal is to build a self-sustainable domain-specific Ontology for the purposes of creating a Knowledge Search Engine. We focused to build it in the Marathi language, which will help school-going children to explore science-related terms. For this, a method is proposed, in which ontology is learned automatically using deep learning model, Bidirectional Long Short-Term Memory (BiLSTM). This paper proposes to use learned ontology to retrieve domain-specific knowledge. The knowledge search engine, which uses constructed ontology to displays search results in Marathi with a very strict limit to the Knowledge complexity of the search results. Unlike, standard search engines, our engine attempts to provide learning resources directly to the user rather than website links. This approach enables the user to directly get information without having to spend time browsing indexed links.