Bayesian Neural Networks in Prediction of Geomagnetic Storms

Andrejková, Gabriela

doi:10.1142/9789812562531_0016

Cited by 4 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The neuro-fuzzy network is a fuzzy logic system into neural networks [Buckley and Hayashi, 1994]. Andrejková et al [1997] applied a fuzzy ANN type to predict geomagnetic storms. Also, Sharifi et al [2006a] introduced a method to estimate Dst index based on an analysis of singular spectrum for extracting the main components with a locally linear neuro-fuzzy network.…”

Section: Introductionmentioning

confidence: 99%

Forecasting the Dst index using a swarm‐optimized neural network

et al. 2017

View full text Add to dashboard Cite

A hybrid technique that combines an artificial neural network with a particle swarm optimization (ANN+PSO) was used to forecast the disturbance storm time (Dst) index from 1 to 6 h ahead. Our ANN was optimized by PSO to update ANN weights and to predict the short‐term Dst index using past values as input parameters. The database used contains 233,760 hourly data from 1 January 1990 to 31 August 2016, considering storms and quiet period, grouped into three data sets: learning set (with 116,880 hourly data points), validation set (with 58,440 data points), and testing set (with 58,440 data points). Several ANN topologies were studied, and the best architecture was determined by systematically adding neurons and evaluating the root‐mean‐square error (RMSE) and the correlation coefficient (R) during the training process. These results show that the hybrid algorithm is a powerful technique for forecasting the Dst index a short time in advance like t + 1 to t + 3, with RMSE from 3.5 nT to 7.5 nT, and R from 0.98 to 0.90. However, t + 4 to t + 6 predictions become slightly more uncertain, with RMSE from 8.8 nT to 10.9 nT, and R from 0.86 to 0.79. Additionally, an exhaustive analysis according to geomagnetic storm magnitude was conducted. In general, the results show that our hybrid algorithm can be correctly trained to forecast the Dst index with appropriate precision and that Dst past behavior significantly affects adequate training and predicting capabilities of the implemented ANN.

show abstract

Section: Introductionmentioning

confidence: 99%

Forecasting the Dst index using a swarm‐optimized neural network

et al. 2017

View full text Add to dashboard Cite

show abstract

“…The evolutionary principle of survival means that individuals "fitted in a better way" have a better chance of surviving and delivering offspring. In the next generation, the number of "good" genes is greater, and the offspring are better "fitted" to the environment (Chodak & Kwaśnicki, 2002;Vandeva, 2012;Awange, et al, 2018;Andrejkova, et al, 2019). The assessment of the adaptation of a given individual to the environment is made using the fitness function.…”

Section: Genetic Algorithmmentioning

confidence: 99%

Genetic Algorithm as Automated Valuation Model Component in Real Estate Investment Decisions System

Chmielewska

Adamiczka²,

Romanowski

2020

Real Estate Management and Valuation

View full text Add to dashboard Cite

Every real-estate related investment decision making process calls for the careful analysis of available information even though it is often carried out in conditions of uncertainty. The paper attempts to minimize the impact of the factor on the quality of real estate investment decisions through the proposal of application of tools based on the simulation of the process of natural selection and biological evolution. The aim of the study is to analyze the potential of methodology based on genetic algorithms (GA) to build automated valuation models (AVM) in uncertainty conditions and support investment decisions on the real estate market. The developed model facilitates the selection of properties adequate to the adopted assumptions, i.e. individuals best suited to the environment. The tool can be used by real estate investment advisors and potential investors on the market to predict future processes and the proper confrontation of past events with planned events. Even though genetic algorithms are tools that have already found particular application on real estate market, there are still areas that need further studies in the case of more effective uses. The obtained results allow for the possibilities and barriers of applying GA to real estate market analyses to be defined.

show abstract

“…Some researchers, e.g., ; Lundstedt and Wintoft (1994); Freeman et al (1993); Detman (1994); Lundstedt (1996, 1997); Andrejkova et al (1998) took solar wind velocity and interplanetary magnetic field parameters as inputs of various neural networks to predict the D st index for hours in advance.…”

Section: Introductionmentioning

confidence: 99%