Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Build Sustainability of Data Sharing Based on Deep Learning in the ESG Environment

Jiang, Han LU (2024) DABN01 20241
Department of Economics
Department of Statistics
Abstract
Against the backdrop of increasing global emphasis on environmental, social, and
corporate governance (ESG), improving the efficiency of ESG data analysis and the
sustainability of its sharing has become particularly important. This study proposes a
comprehensive research framework, first demonstrating the advantages of deep models
for ESG data analysis, and then discussing the combination of ESG data sharing and
blockchain technology, aiming to achieve sustainable development of ESG data sharing. Firstly, we collected and processed ESG data from multiple companies, including annual
reports, third-party ratings, and survey data, and standardized and extracted features. In
terms of ESG data processing, K-nearest neighbor (KNN) models... (More)
Against the backdrop of increasing global emphasis on environmental, social, and
corporate governance (ESG), improving the efficiency of ESG data analysis and the
sustainability of its sharing has become particularly important. This study proposes a
comprehensive research framework, first demonstrating the advantages of deep models
for ESG data analysis, and then discussing the combination of ESG data sharing and
blockchain technology, aiming to achieve sustainable development of ESG data sharing. Firstly, we collected and processed ESG data from multiple companies, including annual
reports, third-party ratings, and survey data, and standardized and extracted features. In
terms of ESG data processing, K-nearest neighbor (KNN) models are used as
benchmarks, while multi-layer perceptrons (MLP) and long short-term memory networks
(LSTM) are adopted due to their excellent ability to handle complex ESG data. Due to
the excellent performance of LSTM in time series analysis, MLP can effectively extract
nonlinear features, thus outperforming KNN in ESG score prediction. Next, we are
considering using blockchain technology to design a decentralized ESG data sharing
platform to ensure data transparency, security, and privacy. This study is based on a comprehensive framework that combines deep learning models
with data sharing platforms. Through deep modeling, dynamic data updates, and
feedback mechanisms, continuously optimize model and platform design to ensure that
data sharing has practical significance for improving corporate ESG. The results of
model analysis can provide specific improvement suggestions for enterprises, forming a
virtuous cycle and promoting continuous improvement of ESG performance (Less)
Please use this url to cite or link to this publication:
author
Jiang, Han LU
supervisor
organization
course
DABN01 20241
year
type
H1 - Master's Degree (One Year)
subject
keywords
ESG Data Analysis, Deep Learning Models, KNN Model, MLP, LSTM, Blockchain Technology, Data Sharing Platform, Sustainable Development, ESG Score Prediction
language
English
id
9155415
date added to LUP
2024-09-24 08:34:10
date last changed
2024-09-24 08:34:10
@misc{9155415,
  abstract     = {{Against the backdrop of increasing global emphasis on environmental, social, and
corporate governance (ESG), improving the efficiency of ESG data analysis and the
sustainability of its sharing has become particularly important. This study proposes a
comprehensive research framework, first demonstrating the advantages of deep models
for ESG data analysis, and then discussing the combination of ESG data sharing and
blockchain technology, aiming to achieve sustainable development of ESG data sharing. Firstly, we collected and processed ESG data from multiple companies, including annual
reports, third-party ratings, and survey data, and standardized and extracted features. In
terms of ESG data processing, K-nearest neighbor (KNN) models are used as
benchmarks, while multi-layer perceptrons (MLP) and long short-term memory networks
(LSTM) are adopted due to their excellent ability to handle complex ESG data. Due to
the excellent performance of LSTM in time series analysis, MLP can effectively extract
nonlinear features, thus outperforming KNN in ESG score prediction. Next, we are
considering using blockchain technology to design a decentralized ESG data sharing
platform to ensure data transparency, security, and privacy. This study is based on a comprehensive framework that combines deep learning models
with data sharing platforms. Through deep modeling, dynamic data updates, and
feedback mechanisms, continuously optimize model and platform design to ensure that
data sharing has practical significance for improving corporate ESG. The results of
model analysis can provide specific improvement suggestions for enterprises, forming a
virtuous cycle and promoting continuous improvement of ESG performance}},
  author       = {{Jiang, Han}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Build Sustainability of Data Sharing Based on Deep Learning in the ESG Environment}},
  year         = {{2024}},
}