In economics, the concept of Social Responsibility (SR) emerged in the late 1970s with the critical words of Milton Friedman. The term SR implies the responsibility obliged towards society from which various institutions derive the benefits. The present work aims to analyze the impact of social responsibility on the technical efficiency of Public Sector Banks (PSBs) with artificial intelligence podiums. It also identifies an important social responsibility indicator inclusive of four dimensions of SR. Mathematically extracting the importance of each parameter related to the efficiency metrics is tedious. Therefore, supervised machine learning algorithms like Random Forest (RF) and XGBoost (XGB) are applied in this study.
Furthermore, banks' effective implementation of SR policy for sustainable development is discussed based on supervised learning. In this study, the impact of 46 social responsibility indicators on the technical efficiency of PSBs is investigated using Machine Learning and non-parametric techniques. Furthermore, the present paper adds to the body of literature by analyzing which indicator of responsibility better influences bank efficiency using a machine learning model. The results revealed that an important indicator impacting efficiency concerning constant and variable return to scale is an area of specialization and background of employees working in PSBs. This showed that PSBs must look towards their work efficiency towards their employees and staff regarding social responsibility.
JEL Classification: B22, C12, G30, G38