Appendix of Explainable AI for coal Hardgrove grindability index prediction: a genetic programming and linear regression approach

This file, provided as an appendix to the paper “Explainable AI for Coal Hardgrove Grindability Index Prediction: A Genetic Programming and Linear Regression Approach,” contains two main parts that offer detailed background information on the prediction models for the Coal Hardgrove Grindability Index (HGI) and related coal sample statistics.

1. Summary of Regression Models (Appendix A):
This section summarizes various regression models from multiple publications used for HGI prediction, including information about the datasets from different countries (such as Serbia, Poland, India, China, Turkey, and Kentucky).
2. Statistical Analysis and Visualization of the CAER Dataset (Appendix B):
This part provides a detailed statistical description of 902 coal samples from the CAER (University of Kentucky Center for Applied Energy Research (CAER) dataset).
The variables include moisture, ash, volatile matter, carbon, hydrogen, nitrogen, oxygen, sulfur, various mineral components, and other coal properties (such as different organic matter contents and reflectance indices).
The table presents the minimum value, maximum value, mean, and standard deviation for each variable, enabling users to gain a comprehensive understanding of the physical and chemical properties of the coal samples. This information is crucial for developing regression models related to coal grindability prediction.
In addition to the statistical table, we have also included boxplots for the CAER dataset to visually illustrate the distribution and variability of key variables.

Additional Information:
In addition to the tabular data and boxplot figures, related literature references are provided, allowing users to trace the original studies and methods, thereby enhancing the reliability and reproducibility of the dataset.