Classification of slope stability based on Artificial Neural Network and Naive Bayes

Ashanira Mat Deris, . and Badariah Solemon, . (2020) Classification of slope stability based on Artificial Neural Network and Naive Bayes. Journal of Energy and Environment (Malaysia), 12 (2). pp. 1-5. ISSN 1985-7462

Estimating the slope stability is a crucial and critical process as the stability of soil slopes not only depends on the geological factors, but also depends on the physical and topography factors. Due to its challenging process, this study attempts on the prediction of slope stability using machine learning (ML) methods which are ArtificialNeural Network (ANN) and Naive Bayes (NB) classifier using the historical slopecases worldwide. The prediction models were developed based on six input factors namely “unit weight, internal friction angle, cohesion, slope angle, slope height and pore pressure ratio” and factor of safety (FOS) as the output factor. The slope data was collected from the previous studies and divided into 70% training and 30%testing datasets for both models. The classification process of ANN and NB were implemented using python programming and the result shows that ANN prediction model gives better prediction result with accuracy of 95%, compared to NB with 84%of accuracy. The prediction of slope stability is one of the critical interests during the slope design process. Hence, this study may served as a benchmark study for the application of ANN and NB machine learning methods in predicting slope stability.

Full text available from: Official URL: https://journal.uniten.edu.my/index.php/jee/articl...

Abstract

Estimating the slope stability is a crucial and critical process as the stability of soil slopes not only depends on the geological factors, but also depends on the physical and topography factors. Due to its challenging process, this study attempts on the prediction of slope stability using machine learning (ML) methods which are ArtificialNeural Network (ANN) and Naive Bayes (NB) classifier using the historical slopecases worldwide. The prediction models were developed based on six input factors namely “unit weight, internal friction angle, cohesion, slope angle, slope height and pore pressure ratio” and factor of safety (FOS) as the output factor. The slope data was collected from the previous studies and divided into 70% training and 30%testing datasets for both models. The classification process of ANN and NB were implemented using python programming and the result shows that ANN prediction model gives better prediction result with accuracy of 95%, compared to NB with 84%of accuracy. The prediction of slope stability is one of the critical interests during the slope design process. Hence, this study may served as a benchmark study for the application of ANN and NB machine learning methods in predicting slope stability.

[error in script]

Item Type:	Article
AGROVOC Term:	slope
AGROVOC Term:	landslides
AGROVOC Term:	data collection
AGROVOC Term:	classification systems
AGROVOC Term:	geological hazards
AGROVOC Term:	soil properties
AGROVOC Term:	risk assessment
AGROVOC Term:	stability
Geographical Term:	Malaysia
Depositing User:	Nor Hasnita Abdul Samat
Date Deposited:	21 Nov 2025 09:42
Last Modified:	21 Nov 2025 09:42
URI:	http://webagris.upm.edu.my/id/eprint/1558

Actions (login required)

View Item