Decoding Work-Life Balance Conundrum
S. A. Sajidha, Deepshika Vijayanand, Madhumitha Rajagopal, Dhanush Sambasivam, V. M. Nisha · Advances in business information systems and analytics book series · 2024
To unravel the complex facets of optimal equilibrium between professional commitments and personal life, this research delves into the intricacies of work-life balance dynamics, employing machine learning to predict factors influencing equilibrium. A stacking-ensemble method utilizes the mutual complementary effects of the base models to improve the performance with better generalization of the model. ML algorithms including gradient boosting algorithm, ridge regression, lasso regression, huber regression, sgd regressor, support vector regressor (SVR), k neighbours regressor, kernel ridge regressor, RANSAC regressor, K means boosting algorithm, and generalized additive model (GAM) were trained meticulously. Fine-tuning the performance of random forest performance having R2 value of 0.9353, the lowest using the stacking ensemble to 0.9999 was performed. The base model combinations encompass a wide array of techniques with random forest as meta learner gave is a notable improvement in its prediction performance.