Human Resource Analytics on Data Science Employment Based on Specialized Skill Sets with Salary Prediction

Abstract

The research aims to perform meaningful human resource analysis on data science employment using the strong influences of specialized skills set with assisting salary prediction. With explosive big data development, a data science job shortage has occurred with high accurate recruitment demand to hire suitable professionals for specific data science roles. To achieve such outcomes, the current data science employment trends were analyzed based on a secondary dataset. Useful analytics insights for job securement and better career development were provided through the main dashboard. Besides, the significant in-demand data science skill variables were also identified for further effective model building. Particularly, certain data pre-processing techniques were performed extensively to prepare and optimize the dataset for the mentioned human resource analytics purposes. The ensemble model was selected as the most suitable salary prediction model with the lowest Average Squared Error (ASE) on validation. Despite the low prediction accuracy caused by numerous filtered skill variables, the salary prediction model’s main objective was to interpret the relationships between input variables and the target salary levels variable. Overall, the results from both the human resource analytic dashboard and salary prediction model were tally where a detailed analytic report was provided to encourage different data science roles with specific and effective career development guidance, using salary as the motivation key.