MACHINE LEARNING APPROACH FOR PREDICTING CORPORATE SOCIAL RESPONSIBILITY PERCEPTION IN UNIVERSITY STUDENTS; [ABORDAGEM DE APRENDIZADO DE MÁQUINA PARA PREVENIR A PERCEPÇÃO DE RESPONSABILIDADE SOCIAL CORPORATIVA EM ESTUDANTES UNIVERSITÁRIOS]

Lillo-Viedma F.; Severino-González P.; Rodríguez-Quezada E.; Arenas-Torres F.; Sarmiento-Peralta G. · Interciencia (2023)

Resumen

Corporate Social Responsibility has become an important corporate principle. Perception about the use of this concept is regarded by corporate stakeholders as strategically crucial. The present work explores the use of machine learning models to analyze connections between socio-demographic traits and CSR perception. Three models are tested based on information provided by university students: a Neural Network (NN), Random Forest (RF) and a Gradient Boosted Tree model (GBT). These models consider socio–demographic and perception scores as inputs and output features, respectively. Results indicates that the GBT model makes better prediction about perceptions. Furthermore, the RF model estimates feature importance which shows the income level feature as a main predictor of CSR–perception.

Instituciones: Universidad Nacional Mayor de San Marcos