A Decision Tree for Rockburst Conditions Prediction

A Decision Tree for Rockburst Conditions Prediction

Dominic Owusu-Ansah¹, Joaquim Tinoco^1,*, Faramarzi Lohrasb², Francisco Martins¹ and José Matos¹

¹Department of Civil Engineering, University of Minho, ISISE, ARISE, 4800-058 Guimaraes, Portugal

²Department of Mining Engineering, Isfahan University of Technology, Isfahan 84156-83111, Iran

Doi:10.3390/app13116655

Abstract

This paper presents an alternative approach to predict rockburst using Machine Learning (ML) algorithms. The study used the Decision Tree (DT) algorithm and implemented two approaches: (1) using DT model for each rock type (DT-RT), and (2) developing a single DT model (Unique-DT) for all rock types. A dataset containing 210 records was collected. Training and testing were performed on this dataset with 5 input variables, which are: Rock Type, Depth, Brittle Index (BI), Stress Index (SI), and Elastic Energy Index (EEI). Other ML algorithms, such as Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), and GradientBoosting (AdaboostM1), were implemented as a form of comparison to the DT models developed. The evaluation metrics and relative importance were utilized to examine some characteristics of the DT methods. The Unique-DT model showed a promising result of the two DT models, giving an average of (F1 = 0.65) in rockburst condition prediction. Although RF and AdaboostM1 (F1 = 0.66) performed slightly better, Unique-DT is recommended for predicting rockburst conditions because it is easier, more effective, and more accurate.

Keywords: Rockburst; rockburst condition; decision tree; machine learning algorithms; predictions; metric