GX & Decarbonization Research

(著者不明)

This study explores the use of LLMs for ESG evaluation of information companies, focusing on extracting and analyzing ESG-related information through natural language processing. It develops evaluation indicators tailored to the characteris…

(著者不明)

This paper proposes a method for automatically generating improvement proposals for environmental activities in companies. By leveraging AI technology, it aims to derive efficient and effective environmental improvement measures, expecting …

(著者不明)

This paper proposes a method using BERT to extract ESG-related sentences from Japanese securities reports, enabling automated ESG disclosure analysis.

(著者不明)

This paper applies text mining to integrated reports of major real estate developers to extract and analyze characteristics of corporate strategy. It quantitatively captures ESG-related trends in the reports and highlights differences withi…

Kokubu, Hiroyuki

This paper proposes a disclosure-based method for measuring non-financial evidence infrastructure maturity in corporate sustainability disclosures. It introduces N4 and N5 scoring axes, ΔN (year-on-year change), four regression-type classif…

Hiroyuki Kokubu

This paper empirically examines the value of reasoning-heavy LLMs for automated ESG narrative scoring, using data from ten Japanese listed firms. Using a four-model consensus design, it finds that the mean score deviation between the reason…

Hiroyuki Kokubu

This paper examines structured GHG disclosure accessibility for 89 major Japanese listed firms using EDINET statutory filings and LLM-assisted extraction from voluntary reports. Only 23 firms had current-year structured GHG tags in EDINET; …

Hao Xu, Xixuan Peng, Xiaodan Jin +2

This study evaluates co-benefits of reducing CO2 and air pollutants from transport. A random forest model predicts PM2.5 concentrations and health benefits are estimated using GEMM. In the best scenario, CO2 emissions peak in 2032 with a 51…

Benneth Oyinna, Zubairu Usman, Aisha Abisoye +2

This study presents a machine learning framework for monitoring and forecasting Nigeria's industrial CO2 emissions to support the 2060 net-zero target. Comparing four models (MLR, Prophet, SVR, Random Forest), MLR performed best (R²=0.978).…

Ibrahim M.

This review paper discusses recent advances in CO2 capture, utilization, and storage (CCUS), with a particular focus on the application of machine learning techniques to optimize processes and the integration of circular economy principles.…

Kai-Robin Lange, José Cassola, Marcelo Lufin +11

This paper applies dynamic topic modeling (RollingLDA) to 1,996 validated news articles from 2011-2025 to analyze public narratives around socio-environmental conflicts related to Chile's energy transition. It identifies twelve topics, reve…

Gyu Jin Pyo, Mi Jin Noh, Yang Sok Kim

This paper applies DirectLiNGAM, a causal structure learning algorithm, to analyze vehicle energy consumption in the EV-ICEV transition market. It identifies causal relationships among factors affecting energy consumption, contributing to e…

Alireza Vafaei Sadr, Seyyed Sina Hejazian, Ajith Vemuri +4

This study estimates CO2 equivalent emissions from AI-assisted stroke imaging in the US. The Ideal AI scenario emits ~16,375 metric tons CO2eq/year, and relocating computation to cleaner-energy states reduces emissions by 54.75%, highlighti…

Xiaodan Qi, Yuxin Chen, Hongyan Zhao +1

This study uses explainable machine learning (XGBoost with SHAP) to predict sustainable purchase intention for green prepared dishes based on 805 survey responses from Jilin Province, China. Predictors are organized into environmental cogni…

Abdulkarim Hamdan J. Alhazmi, Sardar M. N. Islam, M. Prokofieva

Proposes the CMA Agentic AI Platform to address three governance challenges in Saudi Arabia's audit market. Segment 1 uses autonomous drone swarms for asset verification and ESG compliance monitoring via deep learning and thermal imaging. S…

Delia Deliu

This study explores how digital transformation and sustainability transition (Twin Transition) reshape auditing. It finds that Agentic AI, blockchain, and big data enable continuous auditing and ESG traceability but raise issues of algorith…

Mohamed Sheta, Bassam A. Ibrahim, M. Osman +2

This study uses machine learning to examine how internal audit competency and maturity affect ESG assurance engagement. Findings show that competency is the most influential factor, outweighing structural maturity. The model explains 50.97%…

Ahyahudin Sodri, G. B. Imasuly, Nuraeni Nuraeni +1

This study develops an explainable machine learning framework using XGBoost and SHAP to assess urban flood vulnerability across Indonesia. Integrating satellite and geospatial data, it computes a Flood Vulnerability Index (FVI) for 514 dist…

Behzad Satari

This study proposes a comprehensive framework integrating life cycle assessment (LCA), techno-economic analysis (TEA), and machine learning (ML) for sustainability assessment of hydrothermal carbonization (HTC) of food and agro-industrial w…

Ahmed Seffah

This paper argues that genuine sustainability requires organizational transformation over mere technology adoption, proposing a human-centric AI framework. It emphasizes leadership and strategic use of AI for sustainability goals.