XGBSO: An Employability Assessment Model

Abstract

To meet the increasing competition in the marketplace during the post epidemic era, it is crucial for the IT industry to swiftly assess and select job seekers while identifying the enhancement skills required for individual job seekers. Current research lacks model optimization and predominantly uses a combination of machine learning methods. However, this study addresses this gap by employing multiple machine learning algorithms based on annual developer survey data from Stack Overflow to predict job seeker skill judgments and quantitative feature scores. The optimized XGBoost model based on the snake optimization algorithm (XGBSO) shows superior performance in accuracy and F1 score compared to other models, accurately predicting the job search ability of programmers. Feature weighting analysis reveals that computer skill is the most crucial feature. The XGBSO model is validated through ROC curves and AUC-ROC values, displaying an accuracy of 0.838 and an F1 score of 0.855, thus indicating excellent performance. The proposed XGBSO model serves as an effective tool for assessing programmers' job search ability and yields significant results.