Exploring the Relationship Between Machine Learning, Good Governance, and Organizational Performance in the Moroccan Public Secto

Sažetak

This article is part of a rigorous exploration of the impact of good governance
principles on the performance of Moroccan public organizations, mobilizing
advanced Machine Learning methods such as XGBoost, CNN and Random
Forests. The results highlight a statistically significant correlation between prac
tices such as transparency, accountability and leadership, and the operational
effectiveness of public institutions. Transparency and accountability stand out
for their decisive influence, with particularly low p-values, underlining their fun
damental role in structuring organizational performance. By guaranteeing more
informed and accountable management of public resources, these principles not
only optimize efficiency but also strengthen the legitimacy of institutions in the
eyes of citizens. By adopting an integrated approach to governance, combining
leadership, employee motivation and sustainable development, the study reveals
that these levers contribute to the continuous improvement of organizational per
formance, although their effects are sometimes more diffuse and sensitive to the
institutional context. The strength of this analysis lies in its ability to demon
strate that Machine Learning algorithms capture complex dynamics, often im
perceptible to the naked eye, and thus offer powerful predictive tools for public
decision-makers. This work opens new perspectives on how reforms can be an
ticipated and adjusted to maximize the impact of public policies. As such, this
study not only enriches the literature on public governance but also proposes an
innovative framework for strengthening strategic decision-making capabilities in
the public sector.