This study compares the landslide susceptibility maps from four application models, namely, (1) the bivariate model of the Dempster-Shafer based evidential belief function (EBF); (2) integration of the EBF in the knowledge-based analytical hierarchy process (AHP) as a pairwise comparison model processed by using all available causative factors; (3) integration of the EBF in the knowledge-based AHP as a pairwise comparison model by using high nominated causative factor weights only; and (4) integrated EBF in the logistic regression (LR) as a multivariate model by using nominated causative factor weights only. These models were tested in Pohang and Gyeongju Cities (South Korea) by using the geographic information system GIS platform. In the first step, a landslide inventory map consisting of 296 landslide locations were prepared from various data sources. Then, a total of 15 landslide causative factors (slope angle, slope aspect, curvature, surface roughness, altitude, distance from drainages, stream power index, topographic wetness index, wood age, wood diameter, wood type, forest density, soil thickness, soil texture, and soil drainage) were extracted from the database and then converted into a raster. Final susceptibility maps exhibit close results from the two models. Models 1 and 3 predicted 82.3% and 80% of testing data during the analysis, respectively. Thus, Models 1 and 3 show better performance than LR. These resultant maps can be used to extend the capability of bivariate statistical based model, by finding the relationship between each single conditioning factor and landslide locations, moreover, the proposed ensemble model can be used to show the interrelationships importance between each conditioning factors, without the need to refer to the multivariate statistic. The research outcome may provide powerful tools for natural hazard assessment and land use planning.
An ensemble algorithm of data mining decision tree (DT)-based CHi-squared Automatic Interaction Detection (CHAID) is widely used for prediction analysis in variety of applications. CHAID as a multivariate method has an automatic classification capacity to analyze large numbers of landslide conditioning factors. Moreover, it results two or more nodes for each independent variable, where every node contains numbers of presence or absence of landslides (dependent variable). Other DT methods such as Quick, Unbiased, Efficient Statistic Tree (QUEST) and Classification and Regression Trees (CRT) are not able to produce multi branches based tree. Thus, the main objective of this paper is to use CHAID method to perform the best classification fit for each conditioning factors, then, combined it with logistic regression (LR) to find the corresponding coefficients of best fitting function that assess the optimal terminal nodes. In the first step, a landslide inventory map with 296 landslide locations were extracted from various sources over the Pohang-Kyeong Joo catchment (South Korea). Then, the inventory was randomly split into two datasets, 70 % was used for training the models, and the remaining 30 % was used for validation purpose. Thirteen landslide conditioning factors were used for the susceptibility modeling. Then, CHAID was applied and revealed that some conditioning factors such as altitude, soil drain, soil texture and TWI, as terminal nodes and reflected the best classification fit. Then, a proposed ensemble technique was applied and the interpretations of the coefficients showed that the relationship between the decision tree branch nodes distance from drain, soil drain, and TWI, respectively, leads to better consequences assessment of landslides in the current study area. The validation results showed that both success and prediction rates, 75 and 79 %, respectively. This study proved the efficiency and reliability of ensemble DT and LR model in landslide susceptibility mapping.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.