Wyniki wyszukiwania - BazTech

1

GIS-based landslide susceptibility assessment and mapping in Ajloun and Jerash governorates in Jordan using genetic algorithm-based ensemble models

Mabdeh Ali Nouh, Al Fugara A’kif, Ahmadlou Mohammad, Al Adamat Rida, Al Shabeeb Abdel Rahman

Acta Geophysica

|

2022

|

Vol. 70, no 3

1253--1267

EN

Landslides are a geological phenomenon that is causing considerable economic and human losses annually in various regions of the world. In some cases, the complex behaviors of some such phenomena cause that single machine learning models fail in modeling them well. To overcome this issue, this paper presents two novel genetic-algorithm (GA)-based ensemble models constructed with the decision tree (DT), k-nearest neighbors (KNN), and Naive Bayes (NB) models based on the bagging and random sub-space (RS) methods for landslide susceptibility assessment and mapping in Ajloun and Jerash governorates in Jordan. Sixteen factors, including topographic, climatic, human, and geological factors were used as possible factors that influence landslide occurrence in the study area. In addition to this, one hundred and ninety two landslide locations were employed for training and testing the models. The GA was used in this study for feature selection based on three models: DT, KNN, and NB. Model performance evaluation based on the area under the receiver operating characteristic (AUROC) curve indicated that the ensemble models outperform the standalone ones. The values of the AUROC curves in the validation phases for the five models, namely, the GA-based DT, KNN, NB, bagging-based, and RS-based ensemble model, were 0.63, 0.69, 0.63, 0.89, and 0.95, respectively. The results of this study suggest that simple models can be combined using the bagging and RS methods to produce integrated models that have higher accuracy than that of any of the individual simple models.

2

Drought forecasting using new advanced ensemble-based models of reduced error pruning tree

Shahdad Mojtaba, Saber Behzad

Acta Geophysica

|

2022

|

Vol. 70, no 2

697--712

EN

The present study investigates the prediction accuracy of standalone Reduced Error Pruning Tree model and its integration with Bagging (BA), Dagging (DA), Additive Regression (AR) and Random Committee (RC) for drought forecasting on time scales of 3, 6, 12, 48 months ahead using Standard Precipitation Index (SPI), which is among the most common criteria for testing drought prediction, at Kermanshah synoptic station in western Iran. To this end, monthly data obtained from a 31-year period record including rainfall, maximum and minimum temperatures, and maximum and minimum relative humidtty rates were considered as the required input to predict SPI. In addition, different inputs were combined and constructed to determine the most effective parameter. Finally, the obtained results were validated using visual and quantitative criteria. According to the results, the best input combination comprised both meteorological variable and SPI along with lag time. Although hybrid models enhanced the results of standalone models, the accuracy of the best performing models could vary on different SPI time scales. Overall, BA, DA and RC models were much more effective than AR models. Moreover, RMSE value increased from SPI (3) to SPI (48), indicating that performance modeling would become much more challenging and complex on higher time scales. Finally, the performance of the newly developed models was compared with that of conventional and most commonly used Support Vector Machine and Adaptive Neuro-Fuzzy Inference System (ANFIS) models, regarded as the benchmark. The results revealed that all the newly developed models were characterized by higher prediction power than ANFIS and ANN.

3

A Novel Ensemble Model - The Random Granular Reflections

Artiemjew Piotr, Ropiak Krzysztof

Fundamenta Informaticae

|

2021

|

Vol. 179, nr 2

183--203

EN

One of the most popular families of techniques to boost classification are Ensemble methods. Random Forests, Bagging and Boosting are the most popular and widely used ones. This article presents a novel Ensemble Model, named Random Granular Reflections. The algorithm used in this new approach creates an ensemble of homogeneous granular decision systems. The first step of the learning process is to take the training system and cover it with random homogeneous granules (groups of objects from the same decision class that are as little indiscernible from each other as possible). Next, granular reflection is created, which is finally used in the classification process. Results obtained by our initial experiments show that this approach is promising and comparable with other tested methods. The main advantage of our new method is that it is not necessary to search for optimal parameters while looking for granular reflections in the subsequent iterations of our ensemble model.

4

Ensembling a linear regression model with an error mitigation component

Nowosielski A., Kowalski P. A., Kulczycki P.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2018

|

Vol. 12, No. 1

40--43

EN

This paper presents a proposal of a model error mitigation technique based on the error distribution analysis of the original model and creatng the additional model that tempers the error impact in particular domain areas identified as the most sensitive. both models are then combined into single ensemble model. The idea is demonstrated on the trivial two-dimensional linear regression model.