No abstract

Convolutional neural network (CNN) is a widely used method in solving classification and regression applications in industries, engineering, and science. This study investigates the optimizing capability of a swarm intelligence algorithm named moth flame optimizer (MFO) for the optimal search of a CNN hyper-parameters (values of filters) and weights of fully connected layers. The proposed model was run with a 3-dimensional dataset (7 width × 7 height × 12 depth), which was constructed through including seven neighbor pixels (vertically and horizontally) from landslide location and 12 predictor variables. Muong Te district, Lai Chau province, Vietnam was selected as the case study, as it had recently undergone severe impacts of landslides and flash floods. The performance of this proposed model was compared with conventional classifiers, i.e., Random forest, Random subspace, and CNN-optimized Adaptive gradient descend, by using standard metrics. The results showed that the CNN-optimized MFO (Root mean square error = 0.3685, Mean absolute error = 0.2888, Area under Receiver characteristic curve = 0.889 and Overall accuracy = 80.1056%) outperformed the benchmarked methods in all comparing indicators. Besides, the statistical test of difference was also carried out by using the Wilcoxon signed ranked test for non-parametric variables. With these statistical measurements, the proposed model could be used as an alternative solution for landslide susceptibility mapping to support local disaster preparedness plans. INDEX TERMS Convolutional neural network, meta-heuristic algorithm, moth flame optimization algorithm, landslide susceptibility.

No abstract

No abstract

This study’s main objective is to propose a hybrid machine learning model based on a gradient boosting algorithm named LightGBM and an artificial ecosystem-based optimization to improve the accuracy of forest fire susceptibility assessment. Four hundred twenty-six historical forest fires from the NASA portal and thirteen conditional factors including elevation, aspect, slope, curvature, normalized difference vegetation index, normalized difference water index, distance to residence, distance to road, distance to river, temperature, rain, humidity, and wind were used to train the model. The model performance was evaluated and compared with other benchmark models using root mean square error, area under receiver operating characteristics (AUC), and overall accuracy. The results show that the proposed model (AUC = 0.9705) outperforms others, such as Random Forest (AUC = 0.958), AdaBoost (AUC = 0.905), Bagging (AUC = 0.945), and Random Subspace (AUC = 0.938), respectively. The final model was interpreted to better understand the most influential factors of forest fire hazards. Study Implications: This study’s main objective is to propose a hybrid machine learning model based on a gradient boosting algorithm called LightGBM and an artificial ecosystem-based optimization to improve the accuracy of forest fire susceptibility assessment. The final model was interpreted using the Shapley (SHAP) values to understand how input factors are related to the fire susceptibility level. This study fills a gap in the research literature, searching for optimal algorithms to improve forest fire susceptibility mapping.