Smart Human Resource Management System to Maximize Productivity

Abstract

Human resource is one of the most valuable assets in an organization. They are bounded to develop the unique and dynamic aspects that strengthen their competitive advantage to persist in an always changing market environment. In order to recruit a quality candidate for an organization, reducing human involvement and verifying details of the candidate is important in recruitment process. Furthermore, having an idea about how well or poor the employees perform, and how likely the employee attrition can occur is vital in human resource management process. This paper is an attempt to introduce smart human resource management system that can maximize the productivity of an organizational environment using machine learning and blockchain technologies. The end goal of this research is a smart human resource management system that reduces human judgment, time in the candidate selection process and predicts employee performance and attrition to motivate current employers to maximize productivity with minimal financial loss in the workplace environment. Skill assessment and resume classification have been done using unsupervised learning algorithms and natural language processing after extracting raw data from employee resumes using Object Character Recognition. Candidate details verification is done by comparing the hashes of the records which are stored in the blockchain. Employee performance and attrition are predicted using supervised machine learning classification techniques with high accuracy and the result of the final performance is generated as a score for each employee considering the multiple attributes that has been standardized and regulated by some specifically considered e-competence frameworks.