A Novel Integrated Approach of Relevance Vector Machine Optimized by Imperialist Competitive Algorithm for Spatial Modeling of Shallow Landslides

摘要: This research aims at proposing a new artificial intelligence approach (namely RVM-ICA) which is based on the Relevance Vector Machine (RVM) and Imperialist Competitive Algorithm (ICA) optimization for landslide susceptibility modeling. A Geographic Information System (GIS) spatial database was generated from Lang Son city in province (Vietnam). GIS includes inventory map fourteen conditioning factors. The suitability of these factors modeling study area verified by Gain Ratio (IGR) technique. prediction model RVM-ICA established training phases. predictive capability evaluated calculations sensitivity, specificity, accuracy, under Receiver Operating Characteristic curve (AUC). In addition, to assess applicability proposed model, two state-of-the-art soft computing techniques including support vector machine (SVM) logistic regression (LR) were used as benchmark methods. results this show that with AUC = 0.92 achieved high goodness-of-fit both testing datasets. outperformed those SVM 0.91 LR 0.87. experimental confirm newly very promising alternative assist planners decision makers task managing prone areas.