An automated sustainability assessment model: extraction, classification and evaluation of corporate reports using NLP techniques
NLP技法を用いた持続可能性評価モデル：企業報告書の抽出、分類、評価の自動化 (AI 翻訳)
Francisco Javier Rodríguez-Ruiz, Ana María García-Berbaneu, Alexsander Luiz Telpisow-Scheid
🤖 gxceed AI 要約
日本語
本研究は、NLP技術を用いて企業のサステナビリティ報告書からESG関連情報を自動抽出・分類・評価する手法を提案する。欧州の中小企業向けサステナビリティ報告基準に準拠したタクソノミーに基づき、スケーラブルで再現性のある評価を実現する。AI駆動型手法が開示の一貫性と信頼性を向上させる可能性を示す。
English
This study proposes an automated methodology using NLP to extract, classify, and assess ESG content from corporate sustainability reports. It uses a taxonomy aligned with European reporting standards for listed SMEs, enabling scalable and reproducible evaluations. Results highlight the potential of AI-driven methods to enhance consistency and reliability of sustainability assessments.
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
日本ではSSBJ（サステナビリティ基準委員会）が開示基準を策定中であり、本手法は日本企業の報告書評価にも応用可能。特に中小企業向けの基準適合性チェックに有用。
In the global GX context
This paper demonstrates an NLP-based assessment framework aligned with European standards, which could be adapted for global frameworks like ISSB or SEC climate rules. It offers a scalable approach for stakeholders to evaluate disclosure completeness.
👥 読者別の含意
🔬研究者:Provides a practical NLP pipeline for ESG content extraction and classification, advancing automated sustainability assessment research.
🏢実務担当者:Useful for corporate sustainability teams to automate report review and ensure alignment with disclosure standards.
🏛政策担当者:Offers insights into how AI can support regulatory monitoring of ESG disclosures.
📄 Abstract（原文）
Corporate sustainability reports frequently include extensive qualitative information related to Environmental, Social, and Governance (ESG) factors, often presented in unstructured formats and with limited comparability. This lack of standardization hampers the ability of stakeholders to evaluate the completeness, transparency, and alignment of disclosures with regulatory frameworks. To address this challenge, the present study proposes an automated methodology that integrates web scraping and Natural Language Processing (NLP) techniques to extract, classify, and assess ESG-related content from corporate reports. The approach is based on a predefined taxonomy aligned with the Draft European Sustainability Reporting Standards for Listed SMEs and enables scalable, reproducible evaluation of disclosure levels across key ESG dimensions. The results provide insights into reporting practices and highlight the potential of AI-driven methods to enhance the consistency and reliability of sustainability assessments.
🔗 Provenance — このレコードを発見したソース
- openaire https://doi.org/10.4995/carma2025.2025.20580first seen 2026-06-11 05:11:20
🔔 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。