GX & Decarbonization Research

Chandan Sharma, Priya Rani

This study applies NLP and clustering to analyze ESG reporting evolution in India from 2017-2024. It finds ownership structure significantly influences reporting, with private firms more transparent. Firm size affects only environmental dim…

Tao Li, Yi Wu, Mingyang Li

This study proposes a stacking ensemble model with SHAP interpretability to map forest carbon density and net primary productivity. Using forest data from China, it provides insights for management zoning. The work fits the AI×ESG intersect…

An Thi Binh Duong, Linh Tran Khanh Do, Scott McDonald +4

This study develops and empirically tests a strategic framework using evolutionary machine learning (EML) to drive sustainable performance in construction firms. Analyzing 213 Vietnamese construction companies, it finds that aligning EML in…

Yufei Xi, Jiangzhang Zhu, Peng Wang +1

Proposes EvoGame-CAKNet, a hybrid framework for carbon price forecasting integrating evolutionary game theory to model multi-agent strategies, multi-head contextual attention for long-range dependencies, and Kolmogorov-Arnold networks for n…

Gaopeng Jiang, Huihui An, Yaling Tian +2

This study analyzes carbon emission reduction drivers and decoupling effects in the transport industry of the Yangtze River Delta region. Using the extended STIRPAT-Ridge model and Tapio decoupling model, it finds that year-end resident pop…

Ramirez, Dely, Muñoz Tabora, Jonathan, Melgar Dominguez, Ozy Daniel

This deposit provides the dataset and R scripts for a study analyzing the decoupling of GDP per capita from CO2 emissions in Honduras (1990-2023). It includes machine learning methods (k-means, Random Forest) and econometric tests (EKC, Tap…

Zhao C.

This review comprehensively covers multiphase metering and intelligent error correction techniques for CO2 transport in CCUS systems. It highlights the effectiveness of AI/ML-based error correction in improving measurement accuracy, and dis…

Yang J.

This study proposes a machine learning approach to predict ESG disclosure quality using board secretaries' characteristics. It analyzes how secretaries' attributes affect disclosure quality, demonstrating the effectiveness of AI-based predi…

O. Zhytkevych

This paper proposes a conceptual hybrid economic-machine learning framework for decarbonization analysis, integrating Pigouvian externality theory with self-organizing maps (SOM) to cluster countries into homogeneous decarbonization systems…

M. Uddin, Md. Samsuzzaman Sobuz, Mohamed Ghalla +5

This study developed a hybrid machine learning model (XGB-HistGB) to predict compressive strength, cost, and CO2 emissions of biochar-incorporated concrete. Using a dataset of nine input parameters, the model achieved high accuracy (R2=0.95…

Jiaran Liu, Yuheng Huang

This study proposes GBMP-LCA, a gradient boosting–based framework for performance prediction and life-cycle optimization of green building materials. It applies LightGBM to multi-source features including material composition, durability, a…

Dilba Rayaru Kandiyil, M. Sadique, Denise Lee +2

This study proposes a hybrid ML-LCA framework to estimate cement embodied carbon using only publicly available macroeconomic data, predicting clinker-to-cement ratio from GDP per capita and other indicators. The Gradient Boosting model is i…

Ruxiao Li, Bo Zhang, Jiayan Dong +1

Using text analysis on annual reports of Chinese A-share listed firms (2010-2023), this study measures corporate biodiversity attention and finds it significantly enhances sustainable development performance. Mechanisms include alleviating …

Anh Bui-Tuyet, Anh Nguyen-Ngoc-Lan, Nhu Hung Duong +1

Vietnam's ESG disclosure is fragmented and report-based data scarce. This study develops an ML pipeline to classify Vietnamese ESG news on the environmental pillar and maps each to GRI criteria via a ChatGPT-based interpretability agent. Be…

Jaehyun Park

This study compares human coding, rule-based, and LLM semantic scoring architectures for ESG disclosure quality assessment using Korean listed firms' sustainability reports from 2020–2021. LLM-based scoring shows greater convergence with hu…

Mohammed W. A. Saleh, Marwan Mansour, Zaid Jaradat +1

This study examines how ESG performance reduces firm risk (leverage) and how AI disclosure strengthens this effect, using 2020-2024 panel data from 1,310 European firms. Robust regressions show AI disclosure acts as a complementary transpar…

H. Jung, Shaopeng Che, Haein Lee

This study constructs ESG disclosure quality indicators using LLM-assisted content analysis of sustainability reports from S&P 500 Energy firms. It finds limited positive associations between individual indicators and S&P Global ESG Scores,…

Anh Nguyen-Ngoc-Lan, Anh Bui-Tuyet, Nhu Hung Duong +1

This study develops an Artificial Neural Network combined with NLP to classify Vietnamese corporate news into Environmental, Social, and Governance pillars. Using TF-IDF and grid search, the Environmental model achieves 90.48% accuracy and …

Tianli Huang, Grace T. R. Lin

Proposes an Interpretable Quantile Temporal Network (IQTN) for the Chinese carbon allowance (CEA) market, integrating feature gating, causal temporal convolution, and non-crossing quantile layers to forecast multi-horizon VaR and CVaR. Achi…

Alick Andrew Sakala, Yu Chen

Proposes RL-ACO, a reinforcement learning (DQN) framework for adaptive consensus optimization in blockchain-based GHG monitoring. By dynamically tuning cluster size, block interval, and alert priority, it achieves 3,625 TPS at 400 validator…