Bridging Data Silos in Corporate Governance: A Hierarchical Stacking Ensemble With Federated Dynamic Aggregation
企業ガバナンスにおけるデータサイロの橋渡し：階層的スタッキングアンサンブルとフェデレーテッド動的集約 (AI 翻訳)
A. Masum, Md. Abul Kalam Azad, Md. Tofael Ahmed Bhuiyan, Md. Abdur Rahman
🤖 gxceed AI 要約
日本語
本研究は、ESG予測におけるデータの異質性、外れ値、プライバシー制約に対処するため、LightGBM、XGBoost、多層パーセプトロンのスタッキングアンサンブルとフェデレーテッド動的加重平均（FedDWA）アルゴリズムを組み合わせたフレームワークを提案。Refinitivデータ（2010-2022）を用いた実証では、集中型モデルがR²=0.9854、連合型でもR²=0.9799と高い性能を示し、SHAPとLIMEによる説明可能性も確保した。
English
This study proposes a framework combining a stacking ensemble of LightGBM, XGBoost, and MLPs with a novel Federated Dynamic Weighted Averaging (FedDWA) algorithm to address data heterogeneity, outliers, and privacy in ESG forecasting. Empirical analysis on Refinitiv data (2010-2022) shows high performance (R²=0.9854 centralized, 0.9799 federated) and explainability via SHAP and LIME.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本のGX文脈では、SSBJ基準への対応や統合報告書におけるESGデータの品質向上が課題となっている。本手法は、異なる企業間でデータを共有せずに高精度なESG評価を可能にするため、国内企業の開示実務や投資家対応に応用が期待される。
In the global GX context
Globally, this paper addresses critical challenges in ESG data aggregation and privacy under frameworks like ISSB and CSRD. The federated approach enables cross-institutional collaboration without compromising data confidentiality, offering a scalable solution for accurate and interpretable ESG assessments.
👥 読者別の含意
🔬研究者:The novel FedDWA algorithm and stacking ensemble with feature passthrough provide a robust methodological contribution for privacy-preserving ESG forecasting.
🏢実務担当者:Enterprises can adopt this framework to enhance ESG score accuracy while complying with data privacy regulations, improving sustainability reporting and investor confidence.
🏛政策担当者:Regulators may consider the implications of federated learning for standardizing ESG data sharing without centralizing sensitive information, supporting disclosure harmonization.
📄 Abstract（原文）
Accurate Environmental, Social, and Governance (ESG) forecasting is pivotal for modern sustainable finance, yet it remains hampered by data heterogeneity, outlier distortions, and strict privacy regulations that inhibit cross‐institutional data sharing. This study addresses these limitations by proposing a comprehensive framework that synergizes advanced ensemble learning with privacy‐preserving federated architectures. First, a centralized “HSE‐ESG” is developed, utilizing a stacking ensemble of LightGBM, XGBoost, and multilayer perceptrons, synthesized by a Bayesian Ridge metalearner with a unique feature passthrough mechanism to capture complex nonlinear dependencies efficiently. Subsequently, to mitigate data leakage risks and address data silos, the framework transitions to “Fed‐ESGNet,” employing a novel Federated Dynamic Weighted Averaging (FedDWA) algorithm that aggregates client updates based on local validation performance rather than traditional sample volume weighting. Empirical analysis using a longitudinal Refinitiv dataset (2010–2022) reveals that the centralized model achieves a remarkable coefficient of determination ( R 2 ) of 0.9854, while the federated approach maintains near‐parity performance R 2 = 0.9799 despite data fragmentation, significantly outperforming standalone baselines. Furthermore, the integration of Explainable AI (XAI) techniques, specifically SHAP and LIME, guarantees granular decision‐making transparency, establishing a robust, scalable, and interpretable paradigm for secure ESG governance assessment.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.1155/acis/8836162first seen 2026-05-15 21:00:00
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。