Integration of Crowdsourcing into Ontology Relation Extraction
Eunike Andriani Kardinata, Nur Aini Rakhmawati · Procedia Computer Science · 2019
Ontology learning is a continuous process that is always being researched and developed. A learning method for one domain may not be applicable to another because of the different characteristics of the data involved. Researchers have been developing various methodologies to build the highest quality of ontology efficiently. As identified in the previous works, one problem which could not be solved my machine alone is the extra-logical errors. These errors can only be identified by human judges and are usually related to the domain of the ontology. In this research, we aim to catalogue available methods, specifically for relation extraction, and the online incremental algorithms which will allow integration of crowdsourcing into ontology learning—to handle said challenge. We also briefly discussed an existing ontology editor called OntoCop, which may be used as a reference for further research. Henceforth, we propose a framework based on our review to improve the current relation extraction method.