Assessment of advanced random forest and decision tree algorithms for modeling rainfall-induced landslide susceptibility in the Izu-Oshima Volcanic Island, Japan

This study analyzed and compared landslide susceptibility models using decision tree (DT), random forest (RF), and rotation forest (RoF) algorithms at Woomyeon Mountain, South Korea. Out of a total of 145 landslide locations, 102 locations (70%) were used for model training, and the remaining 43 locations (30%) were used for validation. Fourteen landslide conditioning factors were identified, and the contributions of each factor were evaluated using the RRelief-F algorithm with a 10-fold cross-validation approach. Three factors, timber diameter, age, and density had no contribution to landslide occurrence. Landslide susceptibility maps (LSMs) were produced using DT, RF, and RoF models with the 11 remaining landslide conditioning factors: altitude, slope, aspect, profile curvature, plan curvature, topographic position index, elevation-relief ratio, slope length and slope steepness, topographic wetness index, stream power index, and timber type. The performances of the LSMs were assessed and compared based on sensitivity, specificity, precision, accuracy, kappa index, and receiver operating characteristic curves. The results showed that the ensemble learning methods outperformed the single classifier (DT) and that the RoF model had the highest prediction capability compared to the DT and RF models. The results of this study may be helpful in managing areas vulnerable to landslides and establishing mitigation strategies.

Spatial landslide susceptibility assessment using machine learning techniques assisted by additional data created with generative adversarial networks

A comparative study of logistic model tree, random forest, and classification and regression tree models for spatial prediction of landslide susceptibility

Landslide susceptibility mapping is an indispensable prerequisite for landslide prevention and reduction. At present, research into landslide susceptibility mapping has begun to combine machine learning with remote sensing and geographic information system (GIS) techniques. The random forest model is a new integrated classification method, but its application to landslide susceptibility mapping remains limited. Landslides represent a serious threat to the lives and property of people living in the Zigui–Badong area in the Three Gorges region of China, as well as to the operation of the Three Gorges Reservoir. However, the geological structure of this region is complex, involving steep mountains and deep valleys. The purpose of the current study is to produce a landslide susceptibility map of the Zigui–Badong area using a random forest model, multisource data, GIS, and remote sensing data. In total, 300 pre-existing landslide locations were obtained from a landslide inventory map. These landslides were identified using visual interpretation of high-resolution remote sensing images, topographic and geologic data, and extensive field surveys. The occurrence of landslides is closely related to a series of environmental parameters. Topographic, geologic, Landsat-8 image, raining data, and seismic data were used as the primary data sources to extract the geo-environmental factors influencing landslides. Thirty-four layers of causative factors were prepared as predictor variables, which can mainly be categorized as topographic, geological, hydrological, land cover, and environmental trigger parameters. The random forest method is an ensemble classification technique that extends diversity among the classification trees by resampling the data with replacement and randomly changing the predictive variable sets during the different tree induction processes. A random forest model was adopted to calculate the quantitative relationships between the landslide-conditioning factors and the landslide inventory map and then generate a landslide susceptibility map. The analytical results were compared with known landslide locations in terms of area under the receiver operating characteristic curve. The random forest model has an area ratio of 86.10%. In contrast to the random forest (whole factors, WF), random forest (12 major factors, 12F), decision tree (WF), decision tree (12F), the final result shows that random forest (12F) has a higher prediction accuracy. Meanwhile, the random forest models have higher prediction accuracy than the decision tree model. Subsequently, the landslide susceptibility map was classified into five classes (very low, low, moderate, high, and very high). The results demonstrate that the random forest model achieved a reasonable accuracy in landslide susceptibility mapping. The landslide hazard zone information will be useful for general development planning and landslide risk management.

Real-time gait trajectory planning is challenging for legged robots walking on unknown terrain. In this paper, to realize a more efficient and faster motion control of a quadrupedal robot, we propose an optimized gait planning generator (GPG) based on the decision tree (DT) and random forest (RF) model of the robot leg workspace. First, the framework of this embedded GPG and some of the modules associated with it are illustrated. Aiming at the leg workspace model described by DT and RF used in GPG, this paper introduces in detail how to collect the original data needed for training the model and puts forward an Interpolation Labeling with Dilation and Erosion (ILDE) data processing algorithm. After the DT and RF models are trained, we preliminarily evaluate their performance. We then present how these models can be used to predict the location relation between a spatial point and the leg workspace based on its distributional features. The DT model takes only 0.00011 s to process a sample, while the RF model can give the prediction probability. As a complement, the PID inverse kinematic model used in GPG is also mentioned. Finally, the optimized GPG is tested during a real-time single-leg trajectory planning experiment and an unknown terrain recognition simulation of a virtual quadrupedal robot. According to the test results, the GPG shows a remarkable rapidity for processing large-scale data in the gait trajectory planning tasks, and the results can prove it has an application value for quadruped robot control.

Landslide susceptibility assessment in Lianhua County (China): A comparison between a random forest data mining technique and bivariate and multivariate statistical models

Developing a smart system for binary classification of disordered voices using machine learning.

Developing metabolomic signatures of diets is a promising strategy to better understand the diet-health paradigm. Our objective was to develop and validate machine learning (ML) models that predict the consumption of a Mediterranean diet (MedDiet) versus a control diet using untargeted metabolomics data from controlled feeding studies. In the Development set, 26 participants (100% men) aged 24-62 years consumed a North American diet for 5 weeks followed by a MedDiet for 5 weeks in full-feeding conditions. In the Validation set, 70 participants (54% men) aged 25-50 years were instructed to follow Canada's Food Guide recommendations for 4 weeks and then consumed a MedDiet for 4 weeks in full-feeding conditions. Plasma metabolites were analyzed using a MPLEx method and an untargeted metabolomics approach. Random forest (RF) and decision tree (DT) models were developed to predict diet assignment using data from the Development set and validated using data from the Validation set. The RF model from the Development set predicted diet assignment with an accuracy of 0.97 (95 %CI: 0.81-1.00). When applied to the Validation set, the RF model had an accuracy of 0.79 (95 %CI:0.71-0.86). Similar results were obtained using the DT model. RF and DT models can predict the consumption of a MedDiet diet with high accuracy in full-feeding conditions in males based on plasma untargeted metabolomics data. However, accuracy is reduced when models are applied to a more heterogenous sample (sample of males and females and less controlled feeding conditions).

BackgroundThis study aimed to investigate the effect of gastrointestinal bleeding (GIB) on the short-term survival of hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) patients, establish a prediction model for HBV-ACLF-related GIB via machine learning (ML) algorithms, and compare the predictive ability of various models.MethodsA total of 583 HBV-ACLF patients from two medical centers were retrospectively enrolled, and patients from one of the centers were randomly divided into a training cohort (n = 360) and a test cohort (n = 153) at a 7:3 ratio. Patients from the other center composed the validation cohort (n = 70). Patients were divided into GIB and non-gastrointestinal bleeding (NGIB) groups according to whether they had GIB during hospitalization, and short-term survival rates were compared between the two groups. Least absolute shrinkage and selection operator (LASSO) regression was used to screen for features associated with GIB. On the basis of the screened features, we used five ML algorithms, namely, logistic regression (LR), support vector machine (SVM), decision tree (DT), random forest (RF), and K-nearest neighbors (KNN), to build a prediction model for GIB. Six metrics, namely, accuracy, area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were used to evaluate the predictive ability of these models.ResultsIn the training cohort, patients in the GIB group had significantly lower 30- and 90-day survival rates than did those in the NGIB group (48.72% versus 85.67% and 10.26% versus 64.80%, respectively), and similar results were obtained in the test cohort and the validation cohort. LASSO regression screened seven features associated with GIB, of which portal hypertension, electrolyte disturbance, and white blood cell counts were modeled features common to the five machine prediction models. The AUCs of the LR, SVM, DT, RF, and KNN models in the training cohort were 0.819, 0.924, 0.661, 1.000, and 0.865, respectively. Compared with the other four models, the LR model had the lowest PPV of 0.202 in the test cohort, the SVM model had the lowest AUC and sensitivity of 0.657 and 0.500 in the validation cohort, the DT model had the lowest sensitivity of 0.436 and 0.438 in the training and test cohorts, respectively, and the KNN model had the lowest PPV of 0.250 in the validation cohort. Notably, the RF model had the least fluctuations in accuracy, AUC, sensitivity, specificity, PPV, and NPV among the 3 cohorts, with good overall predictive ability.ConclusionGIB has a significant effect on short-term survival in patients with HBV-ACLF. On this basis, five ML prediction models, LR, SVM, DT, RF, and KNN, were established to have better prediction ability for GIB, among which the RF model has the most robust prediction performance, which can help clinicians intervene in advance and improve the short-term survival rate of patients.

In this study, hybrid integration of MultiBoosting based on two artificial intelligence methods (the radial basis function network (RBFN) and credal decision tree (CDT) models) and geographic information systems (GIS) were used to establish landslide susceptibility maps, which were used to evaluate landslide susceptibility in Nanchuan County, China. First, the landslide inventory map was generated based on previous research results combined with GIS and aerial photos. Then, 298 landslides were identified, and the established dataset was divided into a training dataset (70%, 209 landslides) and a validation dataset (30%, 89 landslides) with ensured randomness, fairness, and symmetry of data segmentation. Sixteen landslide conditioning factors (altitude, profile curvature, plan curvature, slope aspect, slope angle, stream power index (SPI), topographical wetness index (TWI), sediment transport index (STI), distance to rivers, distance to roads, distance to faults, rainfall, NDVI, soil, land use, and lithology) were identified in the study area. Subsequently, the CDT, RBFN, and their ensembles with MultiBoosting (MCDT and MRBFN) were used in ArcGIS to generate the landslide susceptibility maps. The performances of the four landslide susceptibility maps were compared and verified based on the area under the curve (AUC). Finally, the verification results of the AUC evaluation show that the landslide susceptibility mapping generated by the MCDT model had the best performance.

The logistic regression (LR) and decision tree (DT) models are widely used for prediction analysis in a variety of applications. In the case of landslide susceptibility, prediction analysis is important to predict the areas which have high potential for landslide occurrence in the future. Therefore, the purpose of this study is to analyze and compare landslide susceptibility using LR and DT models by running three algorithms (CHAID, exhaustive CHAID, and QUEST). Landslide inventory maps (762 landslides) were compiled by reference to historical reports and aerial photographs. All landslides were randomly separated into two data sets: 50% were used to establish the models (training data sets) and the rest for validation (validation data sets). 20 factors were considered as conditioning factors related to landslide and divided into five categories (topography, hydrology, soil, geology, and forest). Associations between landslide occurrence and the conditioning factors were analyzed, and landslide-susceptibility maps were drawn using the LR and DT models. The maps were validated using the area under the curve (AUC) method. The DT model running the exhaustive CHAID algorithm (prediction accuracy 90.6%) was better than the DT CHAID (AUC = 90.2%), LR (AUC = 90.1%), and DT QUEST (84.3%) models. The DT model running the exhaustive CHAID algorithm is the best model in this study. Therefore, all models can be used to spatially predict landslide hazards.

Comparison of landslide susceptibility based on a decision-tree model and actual landslide occurrence: The Akaishi Mountains, Japan

GIS-based evaluation of landslide susceptibility using hybrid computational intelligence models

The Mugling–Narayanghat road section falls within the Lesser Himalaya and Siwalik zones of Central Nepal Himalaya and is highly deformed by the presence of numerous faults and folds. Over the years, this road section and its surrounding area have experienced repeated landslide activities. For that reason, landslide susceptibility zonation is essential for roadside slope disaster management and for planning further development activities. The main goal of this study was to investigate the application of the frequency ratio (FR), statistical index (SI), and weights-of-evidence (WoE) approaches for landslide susceptibility mapping of this road section and its surrounding area. For this purpose, the input layers of the landslide conditioning factors were prepared in the first stage. A landslide inventory map was prepared using earlier reports, aerial photographs interpretation, and multiple field surveys. A total of 438 landslide locations were detected. Out these, 295 (67 %) landslides were randomly selected as training data for the modeling using FR, SI, and WoE models and the remaining 143 (33 %) were used for the validation purposes. The landslide conditioning factors considered for the study area are slope gradient, slope aspect, plan curvature, altitude, stream power index, topographic wetness index, lithology, land use, distance from faults, distance from rivers, and distance from highway. The results were validated using area under the curve (AUC) analysis. From the analysis, it is seen that the FR model with a success rate of 76.8 % and predictive accuracy of 75.4 % performs better than WoE (success rate, 75.6 %; predictive accuracy, 74.9 %) and SI (success rate, 75.5 %; predictive accuracy, 74.6 %) models. Overall, all the models showed almost similar results. The resultant susceptibility maps can be useful for general land use planning.

Voice Disorder Classification Using Wav2vec 2.0 Feature Extraction