Answers are grounded in this paper and cite their sources.
<ns5:p> Background The rapid expansion of digital technologies has transformed business ecosystems and increased the importance of integrating Environmental, Social, and Governance (ESG) accounting into sustainable value creation. However, prior studies have largely examined ESG accounting, data analytics, and artificial intelligence separately, with limited evidence regarding their integrated and nonlinear relationships within digital business ecosystems. This study aims to develop a machine learning-based framework to analyze the relationships among ESG accounting, data analytics capability, and sustainable value creation. Methods This study employed a quantitative and data-driven research design using firm-level ESG, financial, and digital capability data. Two machine learning algorithms, Random Forest and Extreme Gradient Boosting (XGBoost), were applied to model complex and nonlinear relationships. Model performance was evaluated using Mean Absolute Error, Root Mean Square Error, and coefficient of determination (R <ns5:sup>2</ns5:sup> ). Feature importance analysis and Shapley Additive Explanations were also used to improve model interpretability and identify key predictors of sustainable value creation. Results The findings show that XGBoost outperformed Random Forest and linear regression models, achieving the highest predictive accuracy (R <ns5:sup>2</ns5:sup> = 0.87). ESG accounting significantly influenced sustainable value creation both directly and indirectly through data analytics capability, confirming its mediating role. Governance and data analytics capability emerged as the most influential predictors of sustainable value creation. Furthermore, nonlinear analysis revealed threshold effects, indicating that ESG initiatives generate substantial value only after reaching a certain maturity level. Conclusions This study demonstrates that sustainable value creation in digital business ecosystems is strongly influenced by the integration of ESG accounting, data analytics capability, and artificial intelligence-driven modeling. The proposed framework contributes to sustainability accounting and digital business research by introducing a nonlinear and interpretable machine learning approach for analyzing ESG-driven value creation. </ns5:p>
No takes yet. Share an insight, caveat, or question.
Kurniawati et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: