Spatio-temporal analysis of forest fire events in the Margalla Hills, Islamabad, Pakistan using socio-economic and environmental variable data with machine learning methods

2021 ◽  
Author(s):  
Aqil Tariq ◽  
Hong Shu ◽  
Saima Siddiqui ◽  
Iqra Munir ◽  
Alireza Sharifi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Forest Fire ◽  
Temporal Analysis ◽  
Environmental Variable ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Variable Data ◽  
Spatio Temporal

scholarly journals Machine Learning Methods for Seismic Hazards Forecast

Geosciences ◽  
2019 ◽  
Vol 9 (7) ◽  
pp. 308 ◽  
Author(s):  
Valeri G. Gitis ◽  
Alexander B. Derendyaev
Keyword(s):  
Machine Learning ◽  
Training Sample ◽  
Time Interval ◽  
Minimum Area ◽  
Learning Methods ◽  
Web Based ◽  
Strong Earthquakes ◽  
Machine Learning Methods ◽  
Regression Approach ◽  
Spatio Temporal

In this paper, we suggest two machine learning methods for seismic hazard forecast. The first method is used for spatial forecasting of maximum possible earthquake magnitudes ( M m a x ), whereas the second is used for spatio-temporal forecasting of strong earthquakes. The first method, the method of approximation of interval expert estimates, is based on a regression approach in which values of M m a x at the points of the training sample are estimated by experts. The method allows one to formalize the knowledge of experts, to find the dependence of M m a x on the properties of the geological environment, and to construct a map of the spatial forecast. The second method, the method of minimum area of alarm, uses retrospective data to identify the alarm area in which the epicenters of strong (target) earthquakes are expected at a certain time interval. This method is the basis of an automatic web-based platform that systematically forecasts target earthquakes. The results of testing the approach to earthquake prediction in the Mediterranean and Californian regions are presented. For the tests, well known parameters of earthquake catalogs were used. The method showed a satisfactory forecast quality.

scholarly journals Natural language processing systems for data extraction and mapping on the basis of unstructured text blocks

2020 ◽  
Vol 26 (3) ◽  
pp. 53-61
Author(s):  
Pavel Kikin ◽  
Alexey Kolesnikov ◽  
Alexey Portnov ◽  
Denis Grischenko
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Mathematical Models ◽  
Optimal Algorithm ◽  
The State ◽  
Gradient Boosting ◽  
Learning Methods ◽  
Data Set ◽  
Machine Learning Methods ◽  
Spatio Temporal

The state of ecological systems, along with their general characteristics, is almost always described by indicators that vary in space and time, which leads to a significant complication of constructing mathematical models for predicting the state of such systems. One of the ways to simplify and automate the construction of mathematical models for predicting the state of such systems is the use of machine learning methods. The article provides a comparison of traditional and based on neural networks, algorithms and machine learning methods for predicting spatio-temporal series representing ecosystem data. Analysis and comparison were carried out among the following algorithms and methods: logistic regression, random forest, gradient boosting on decision trees, SARIMAX, neural networks of long-term short-term memory (LSTM) and controlled recurrent blocks (GRU). To conduct the study, data sets were selected that have both spatial and temporal components: the values of the number of mosquitoes, the number of dengue infections, the physical condition of tropical grove trees, and the water level in the river. The article discusses the necessary steps for preliminary data processing, depending on the algorithm used. Also, Kolmogorov complexity was calculated as one of the parameters that can help formalize the choice of the most optimal algorithm when constructing mathematical models of spatio-temporal data for the sets used. Based on the results of the analysis, recommendations are given on the application of certain methods and specific technical solutions, depending on the characteristics of the data set that describes a particular ecosystem

Statistical-based dynamic machine learning models for nonlinear spatio-temporal processes

2018 ◽  
Author(s):  
◽  
Patrick McDermott
Keyword(s):  
Machine Learning ◽  
Uncertainty Quantification ◽  
Process Model ◽  
Point Estimation ◽  
Dynamical Process ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Statistical Framework ◽  
Temporal Models ◽  
Spatio Temporal

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] One of the most vital aspects of any spatio-temporal model is characterizing the dynamics of the process. In both a spatio-temporal forecasting and data assimilation setting, the dynamical process model often determines the success of a given methodology. Although in some cases mechanistic processes can motivate a dynamical process model, frequently the dynamics involve complex nonlinear behavior that is difficult to specify a priori. This nonlinearity is generally induced by interactions between the process at various spatial locations or between various scales of variability. Over the past few decades, machine learning methods have been shown to be successful for predicting or classifying high-dimensional complex nonlinear multivariate processes. Even though spatio-temporal processes share many of these same characteristics, only recently have machine learning methods been applied to spatio-temporal processes. Furthermore, uncertainty quantification is almost always ignored in the machine learning literature, with a focus only on point estimation prediction or classification. The computational burden of implementing these models within an uncertainty quantification framework often discourages the use of a more traditional statistical framework. Yet, uncertainty quantification is critical for nonlinear spatio-temporal processes where unique behaviors such as non-Gaussianity and extremes are often found. This dissertation focuses on retaining components of machine learning methods for dynamical processes that have been important for their success, while placing them within a formal statistical framework that is computationally feasible. Specifically, a Bayesian framework is employed so the uncertainties associated with the investigated processes can rigorously be quantified. The developed Bayesian machine learning based statistical dynamical spatio-temporal models are shown to outperform a suite of competing methods in terms of both forecast accuracy and uncertainty quantification for multiple nonlinear spatio-temporal processes.

Sign in / Sign up

Export Citation Format

Share Document