Prescriptive Decision Making Model for Contextual Intelligence in Human Resource Analytics

Abstract

This study also proposes Machine learning algorithms, specifically Logistic Regression and Gaussian Naïve Bayes, for generating recommendations; which exploit user context information to shortlist for the desired job role and also recommend alternative jobs to the candidates. Based on existing skills, new opportunities and possibilities will be introduced, that the candidate wouldn't have explored before. In an innovative approach, it also focuses on formalizing the problem of identifying the additional skills, taking into account the employee's existing skills. Performance of the proposed system is evaluated in terms of classification accuracy and the results are compared with alternative models. With an objective to assist job seekers and recruiters in selecting the perfect jobs and the right candidates to achieve career objectives and desired goals, a bidirectional recommender system is introduced later in this research work. This system comprises: Named Entity Recognition (NER), Similarity techniques and text summarization techniques. In an attempt to tackle the problem of unregistered words for text summarization, a solution called Decoder Attention with Pointer Network (DA-PN) has been introduced. This method incorporates the use of a coverage mechanism to prevent word repetition in the generated text summaries. DA-PN + Cover model with mixed learning objective (MLO) (DA-PN + cover + MLO) is utilized for protecting the spread of increasing errors in generated text summaries.