Investigation and analysis of automated job knowledge ontology construction with reduced overhead and time complexity

Abstract

In many industries, the hiring process is a critical task, where many applicants would have applied for a single position, so it is required to analyse their qualification which ought to suit the requirement. Identifying the feasible candidate with the required knowledge and skills is a difficult task which is faced by the organizations. In the earlier work, job-know ontology (JKO) is built, which gives interlink connection among the job, skills and the competence level of the particular job. Nevertheless, this ontology can be utilized to know only the specific job position where various ontologies were required to be built for various jobs. It is very difficult to make the applicants to start their feasible career. The ontology requires updating manually for time being to adjust the existing competence level. newlineIn the first research work, this issue is rectified through domain ontology construction framework (DOCF) from the job knowledge thesis and the job requirements details, which are obtained from online. The obtained documents were clustered regarding the content and conceptual similarity-based clustering method. From those clustered documents job-know ontology will be built. So it is easy to create the information requirements about the job search phase by both candidates and the human resources (HR) of the industries. It is proved that proposed research method leads to give better results when compared with the existing research method in terms of increased accuracy rate. However in this work, memory issues might arise due to the dynamic updation of ontology by retrieving run time details. It can be avoided by finding the similarity between the ontologies. newline