Predicting Anxiety Among Technical Employees: A Machine Learning Approach

Abstract

Amidst the ever-evolving technology landscape, the concerns of technical personnel have taken the forefront, extending their ramifications to job satisfaction and productivity. Anxiety disorder, characterized by feelings of fear and stress, is a prevalent mental health condition among technical employees. In light of this, the presented study in this paper delves into enhancing the accuracy of anxiety prediction. Utilizing proficient machine learning algorithms encompassing Random Forest (RF), Decision Tree (DT), AdaBoost, bagging, Bernoulli Naive Bayes (BNB), Logistic Regression (LR), and Support Vector Machine (SVM), a predictive model is proposed to gauge anxiety levels. The study unveils an impressive accuracy pinnacle of 96.23%, particularly with the implementation of AdaBoost. These findings shed light on the substantive symbiotic relationship between machine learning and anxiety prediction. Ultimately, this study holds pragmatic implications for organizations aiming to fortify employee mental well-being, thereby elevating job satisfaction and productivity within this dynamic technological epoch.