A Review of Machine Learning Ontologies Sinha Prashant Kumar1, Gajbe Sagar Bhimrao2, Chakraborty Kanu3,*, Sahoo Subhranshubhusan4, Debnath Sourav5, Mahato Shiva Shankar6 1Indian Statistical Institute, Documentation Research, and Training Centre, Bangalore, Karnataka 2University of Calcutta, Kolkata, WB 3Indian Institute of Technology (BHU), Varanasi, Uttar Pradesh 4Indian Statistical Institute Documentation Research, and Training Centre Bangalore, Karnataka; University of Calcutta Kolkata, West Bengal 5National Institute of Technology Tiruchirappalli, Tamil Nadu 6Srinivasa Ramanujan Library Indian Institute of Science Education and Research, Pune, Maharashtra *Corresponding Authors: Kanu Chakraborty, kchakraborty.lib@iitbhu.ac.in
Online published on 13 April, 2022. Abstract The research provides an overview of machine learning ontologies (MLOs) based on their purpose, ontology type, design approaches, and other factors. To identify the works a systematic review method was employed and as a consequence, nine papers addressing MLOs were discovered. The bulk of the produced ontologies were domain ontologies with a modular structure, according to the review. Because most of the MLOs were formal ontologies, they can be processed by machines. The web ontology language, a World Wide Web consortium recommended language for ontology representation, has been used to represent the majority of the MLOs. Only a few MLOs acknowledged the development methodology or hierarchy construction process of MLOs and only a handful reused existing vocabularies and ontologies. For the development methodology, ontology development 101 methodology was the preferred choice and for the evaluation of MLOs, task-based evaluation was preferred. Since ontologies were freely available in OWL files, the study includes ontology metrics as well. This study can be used by the research community to better understand the MLO that has been published in the literature, and then use or repurpose these MLOs to meet their objectives or systems. Top Keywords Machine Learning Ontology, Ontology Review, Machine Learning, Ontology, Parameters. Top |