An Ontology-Based Data Mining Approach for Predicting the Research Ideas using Past Research in the Wildlife Sector of Sri Lanka

Abstract

Sri Lanka being a global biodiversity hotspot, places great value for biodiversity owing to ecological, socioeconomic, and cultural factors. However, the wildlife of Sri Lanka is critically threatened due to several factors, mainly human activities and needs dire conservation measures. Inadequate knowledge and technical support also hinder wildlife management activities. Findings of wildlife research studies could be integrated into data-driven conservation and management decisions but the current contribution is not satisfactory. .This research work shows a novel data mining approach for finding hidden keywords and automatic labelling for past research work in this domain. We used Latent Dirichlet Allocation (LDA) algorithms to model topics and identify the major keywords also developed an ontology model to represent the relationships between each keyword. Both approaches are also useful for potential research ideas, to identify research gaps and can classify the subjects related to a publication by non-professional related fields. The experiment results demonstrate the validity and efficiency of the proposed method.