Application of Natural Language Processing Methods to Automated Ontology Building for Domain-Dependent Information Acquisition
R. N. Ashlin Deepa, Bandi Rambabu, G. Senthilkumar, Thresa Jeniffer J, R Surendran · 2025
Ontology is an emerging field with massive potential for improving information organization, management, and understanding. The ontology representation model is utilized for knowledge representation sharing and reusing in Natural Language Processing (NLP). Easy classification of the information is the main feature of the ontology construction. Regardless, there is still an issue in automatically building ontologies from unorganised data. Intelligent association collection from ontologies, hyponymy creation across ontology categories, and dialogue between data providers and domain specialists all contribute to this difficulty. Furthermore, concept, modality and context are the three major dimensions of the knowledge-based ontology construction. The problems concerning ontology construction, mapping, and maintenance are essential areas that must be addressed and focused on. This paper comprehensively reviews the automatic ontology construction methods for predictive analysis. Various automatic ontology construction methods are studied, and the research gap in the existing methods such as Accuracy in prediction, response time and computational complexity are addressed and presented. Based on the literature review, a novel incremental algorithm is proposed for merging the related domain ontologies into a cluster that can strengthen the knowledge base, thereby providing automated inferences and increasing the efficiency of the automated ontology construction.