The accuracy and reliance increase and consequently reduction of uncertainty of spatial prediction maps of environmental hazards including landslides is one of the challenges facing with in such studies Therefore, the objective of this research is to introduce a hybrid model of data mining algorithm named random Forest (RF) -random Subspace (RF-RS) in order to enhance the accuracy of spatial prediction map of landslide prone areas around the city of Bijar, Kurdistan province, Iran. Firstly, 19 affecting factors on shallow landslides in the study area including slope degree, slope aspect, elevation, curvature, profile curvature, plan curvature, solar radiation, stream power index (SPI), topographic wetness index (TWI), length-angle of slope (LS), land use, normalized difference vegetation index (NDVI), litho logy, distance to fault, fault density, rainfall, distance to stream, stream density and distance to road were identified. Then based on Information Gain Ratio (IGR), twelve factors among them were selected to be used in modeling. The elative importance of each factor was assessed by random Forest (RF) model as well as random Forest-random Subspace (RF-RS) hybrid model. Kappa, Precision, Recall, and AUROC indices were used to evaluate the models not only for training dataset but also for testing dataset. Shallow landslide susceptibility maps of the study area were prepared using both models. The results showed that slope aspect in the RF model and slope degree in the RF-RS hybrid model is the most important affecting factor on landslide occurrence in the area. The model evaluation results indicated that both models are reasonable in application for shallow landslide susceptibility mapping. The findings also indicated that the percentage of area under the curve of ROC (AUROC) was 0.729 and 0.784 for training dataset by RF and RF-RS hybrid model, respectively, while these values were 0.717 and 0.771 for testing dataset. In conclusion, it can be claimed that the new technique (RF-RS hybrid model) is able to increase the accuracy of spatial prediction map of shallow landslides in the study area. This accurate map will help decision-makers to protect infrastructures of an area, to develop better land-use planning programs and to more effectively control sediments.

Optimization of electron scattering has been investigated using random potential barriers. random pottential barriers can be defined in two ways: when these line defects are placed on the insulation surface, but strength of their potential is changing randomly, and the other is when potential barriers have a constant value, while their location on the surface of topological insulators is changing randomly. To observe the better passage of electrons, the probability of transmission in the random potential state is calculated N times. These N values are averaged and with the probability of transmission, in the local potential state is compared. It seems that, to propagating of incident electron for some amount of incident energy, the number of defects, strength of potential and even direction of propagation electron, to the same result for the local line defects is close. But for some amounts of incident energy or some structural changes show significant changes.

In this paper, we present the Anderson-Darling (AD) and Kolmogorov-Smirnov (KS) goodness of fit statistics for stationary and non-stationary random fields. Namely, we adopt an easy-to-apply method based on a random projection of a Hilbert-valued random field onto the real line R, and then, applying the well-known AD and KS goodness of fit tests. We conclude this paper by studying the behavior of the proposed approach in the wide range of simulation studies and in a case study of autistic and healthy individuals.

Background: The bagging (BG) and random forest (RF) are famous supervised statistical learning methods based on classification and regression trees. The BG and RF can deal with different types of responses such as categorical, continuous, etc. There are curves, time series, functional data, or observations that are related to each other based on their domain in many statistical applications. The RF methods are extended to some cases for functional data as covariates or responses in many pieces of literature. Among them, random-splitting is used to summarize the functional data to the multiple related summary statistics such as average, etc. Methods: This research article extends this method and introduces the mixed data BG (MD-BG) and RF (MD-RF) algorithm for multiple functional and non-functional, or mixed and hybrid data, covariates and it calculates the variable importance plot (VIP) for each covariate. Results: The main differences between MD-BG and MD-RF are in choosing the covariates that in the first, all covariates remain in the model but the second uses a random sample of covariates.   The MD-RF helps to unmask the most important parts of functional covariates and the most important non-functional covariates. Conclusions: We apply our methods on the two datasets of DTI and Tecator and compare their performances for continuous and categorical responses with the developed R package (“RSRF”) in the GitHub.