Ontology based classification system for online job offers
Ehtesham ul haq Dar, Jürgen Dorn · 2018 International Conference on Computing, Mathematics and Engineering Technologies (iCoMET) · 2018
The significance of employment in a setup of society is quite evident. Publishing jobs online opened the research opportunities to study different methods for the automation of job offers retrieval — from the internet. We need a classification mechanism, from machine learning or some other discipline, to automate job offers retrieval. In this paper, we devised an ontology-based classifier for job offers classification. More than 5000 job offers were collected from different job offers websites. The ontology-based classification works in three stages, concepts are extracted from job offers text description as feature vectors, the minimum threshold for the classification is calculated, and a classification model is developed. Rather using machine learning algorithms, we used an ontology for classification. We evaluated this classifier according to machine learning evaluation model, training, and test dataset. Our classifier showed more than 0.9 accuracy, precision, and recall for both training and test dataset. Our finding paves the way to automate the job offers classification and retrieval.