What is it about?

This study models the relationship between Chinese energy related carbon emissions and its different vital drivers under the ANN and the conventional linear regression. Furthermore, compare the performance of the two models based upon the mean squared errors of the estimates obtained under the two models. This study can work as a reference for building the various kinds of much advanced non-linear ANN models for modelling the relation between a regions carbon emissions and its drivers. Additionally, it can also set the base for future research in comparing the performance of advanced ANN and regression models on the topic under discussion.

Featured Image

Why is it important?

There has been not much work on the relationship between the carbon emissions and their ‘socio-economic’ factors through artificial intelligent (AI) machine learning approaches like the ‘artificial neural networks’. Specifically, for China which is the largest single source of worlds energy based carbon emissions. Furthermore, it’s also very important to answer the not attended important question that: does the ANN model is also superior to regression in predicting the relation between carbon emissions and its drivers? Or does the already available huge volume of literature using linear and advanced regression models is sufficient in modelling the linear or non-linear relationship between the China’s carbon emissions and various driving factors?

Perspectives

This study, contrary to the usual belief in the superiority of complex ANN models over regression analysis, has shown that in some cases the simplest linear regression can produce more accurate results compared to ANN.

Dr. Muhammad Jawad Sajid
Xuzhou Institute of Technology

Read the Original

This page is a summary of: Machine Learned Artificial Neural Networks Vs Linear Regression: A Case of Chinese Carbon Emissions, IOP Conference Series Earth and Environmental Science, June 2020, Institute of Physics Publishing, DOI: 10.1088/1755-1315/495/1/012044.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page