Entropy-Weighted Machine Learning Evaluation of SHG Empowerment in Rural West Bengal
Journal of Intelligent Decision Making and Information Science · 06.10.2026 · 3 (15s) · 1856-1879
DOI: 10.17762/jidmis.v3.4729 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 06.10.2026. Verfasst von Arpan Mukherjee. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Self-Help Groups (SHGs) are promoted as vehicles for women's socio-economic empowerment in India, yet statistically rigorous evidence of measurable impact remains scarce relative to the descriptive literature. This study tests that question using a survey of 3,000 women SHG members from the Birbhum and Burdwan districts of West Bengal, structured around four management pillars — Marketing, Finance, Operations, and Human Resource Development — mapped onto the Life Sustenance, Self-Respect, and Freedom-of-Choice empowerment dimensions. An entropy-weighted composite empowerment index is constructed and benchmarked against a Principal Component Analysis alternative, and the hypothesis that SHG participation yields a significant positive effect (H0) is tested against the null of no effect (Ha) via one-sample and paired t-tests. Five machine-learning models are trained on a 70/15/15 split to assess whether demographic covariates predict empowerment. Composite empowerment scored significantly above the neutral midpoint (t(2999) = 336.09, p < .001, Cohen's d = 6.14), and household income rose 135.3% (paired t(2999) = 81.93, p < .001), supporting H0 over Ha. A Gradient Boosting Regressor explained 68.1% of held-out variance (R2 = 0.681), and a Gradient Boosting Classifier achieved AUC = 0.894 for high- versus low-empowerment members, outperforming linear baselines (R2 = 0.013; AUC = 0.540). Pre-SHG income, savings, and age were the strongest predictors. The study contributes an entropy-weighted indexing method validated against PCA and a reproducible machine-learning framework for evaluating SHG-led empowerment at scale.
Zitieren
APA
Mukherjee, A. (2026). Entropy-Weighted Machine Learning Evaluation of SHG Empowerment in Rural West Bengal. Journal of Intelligent Decision Making and Information Science, 3(15s), 1856-1879. https://doi.org/10.17762/jidmis.v3.4729
Vancouver
Mukherjee A. Entropy-Weighted Machine Learning Evaluation of SHG Empowerment in Rural West Bengal. Journal of Intelligent Decision Making and Information Science. 2026;3(15s):1856-1879. doi:10.17762/jidmis.v3.4729
BibTeX
@article{Mukherjee2026,
author = {Mukherjee, Arpan},
title = {Entropy-Weighted Machine Learning Evaluation of SHG Empowerment in Rural West Bengal},
journal = {Journal of Intelligent Decision Making and Information Science},
year = {2026},
volume = {3},
number = {15s},
pages = {1856--1879},
doi = {10.17762/jidmis.v3.4729}
}
EUVIMED