ABSTRACT Poverty is a major concern in emerging countries like India. The present study predicts the modelling of social efficiency in Indian MFIs employing machine learning, an unexplored area among the extant research. We collected 31 Indian MFIs from the World Bank Mix market database from 2011 to 2019. This research employed eight machine‐learning techniques that include linear regression, random forest, gradient boosting, SVR and KNeighbors regressor. We evaluated the models by applying an 80:20 split through R 2 , MAE and RMSE metrics. The KNeighbors regressor emerged as the relatively performing model with a mean R 2 of 0.5823, exceeding all the models by 8%–27%. Borrowings came out as the prominent predictor elucidating 34%–38% variance in predicting social efficiency, trailed by loan officers. The conclusions offer notable observations to enlighten guidelines and resource allocations for the policymakers for the MFIs to accomplish SDG 1.
Das et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: