scholarly journals Redes Neurais Artificiais na Previsão de Queimadas e Incêndios no Pantanal (Artificial Neural Networks in Prediction of Forest Fires and Burns in the Pantanal)

2017 ◽  
Vol 10 (5) ◽  
pp. 1355
Author(s):  
Hevelyne Henn da Gama Viganó ◽  
Celso Correia de Souza ◽  
Marcia Ferreira Cristaldo ◽  
Leandro De Jesus
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
Time Series ◽  
Artificial Neural Networks ◽  
Forest Fires ◽  
Multilayer Perceptron ◽  
Environmental Variables ◽  
Meteorological Variables ◽  
Burned Area ◽  
Artificial Neural

O bioma pantaneiro é acometido anualmente por um grande número de queimadas e incêndios. Prever esses eventos é de suma importância, uma vez que, prejuízos à fauna e à flora poderiam ser minimizados e catástrofes evitadas. Uma intervenção imediata do poder público na mitigação desses eventos passa, essencialmente, pela previsão do número de focos e das áreas queimadas, e posteriormente, na localização desses focos e na identificação das áreas. Dados precisos sobre as variáveis ambientais, em tempo real, podem ser obtidos através do sensoriamento remoto, aliado aos sistemas de informações geográficas, às técnicas de inteligência artificial e estatística aplicada, favorecendo às tomadas de decisão na previsão. O objetivo deste estudo foi aplicar a técnica de Redes Neurais Artificiais (RNAs) para a previsão dos focos e das áreas queimadas no Pantanal Sul-Mato-Grossense. O centro de previsão de tempo e estudos climáticos do Instituto Nacional de Pesquisas Espaciais (INPE), bem como, o Instituto Nacional de Meteorologia (INMET) possuem bancos de dados do Pantanal, dos quais, as variáveis envolvidas nesse processo foram extraídas. Utilizando as RNAs do tipo Multilayer Perceptron, com algoritmo Retropropagation de aprendizagem foi possível prever o valor do número de focos com um ajuste de 84,8% utilizando um conjunto de variáveis meteorológicas como preditoras e, de 99,4% usando como preditora, somente a série temporal do número de focos. No entanto, o ajuste passou a ser 90,3% ao se realizar a previsão da área queimada, utilizando as mesmas variáveis, e de 98,6%, usando como preditora, somente os dados da área queimada.  A B S T R A C TOn a year basis the Pantanal biome is affected by a large number of fires and fire points. Predicting these events is of paramount importance, since damage to fauna and flora could be minimized and disasters might be avoided. Immediate intervention of the government in mitigating these events essentially depends on the identification and location of fire outbreaks. Accurate and reliable data on environmental variables in real time can be obtained by means of remote sensing coupled with geographic information systems, techniques of artificial intelligence and applied statistics, which would favor the decision-making in foreseeing fire foci and burned area. The aim of this study was to apply the technique of artificial neural networks to predict the fire foci and burned areas in the Pantanal Sul-Mato-Grossense. The center for weather forecast and climate studies of the National Institute for Space Research (INPE) and the National Institute of Meteorology (INMET) afford meteorological database of the Pantanal region, from which the environmental variables involved in this process were extracted. By using Multilayer Perceptron Artificial Neural Networks with Backpropagation algorithm, it was possible to predict the value of the number of fire foci with an adjustment of 84,8% using a set of meteorological variables as predictors and of 99,4% using as a predictor only the time series of the number fire foci.  However, the adjustment became 90,3% when the forecast of the burned area, using the same meteorological variables, and 98,6%, using as a predictor, only the time series of the burned area.Keywords: fire forecasting, environmental monitoring,artificial intelligence. 

scholarly journals Cost Forecasting of Public Construction Projects Using Multilayer Perceptron Artificial Neural Networks: A Case Study

2021 ◽  
Vol 41 (3) ◽  
pp. e87737
Author(s):  
Alcineide Pessoa ◽  
Gean Sousa ◽  
Luiz Maués ◽  
Felipe Alvarenga ◽  
Débora Santos
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Public Sector ◽  
Multilayer Perceptron ◽  
Construction Projects ◽  
Gradient Descent Method ◽  
The Neural Network ◽  
Artificial Neural

The execution of public sector construction projects often requires the use of financial resources not foreseen during the tendering phase, which causes management problems. This study aims to present a computational model based on artificial intelligence, specifically on artificial neural networks, capable of forecasting the execution cost of construction projects for Brazilian educational public buildings. The database used in the training and testing of the neural model was obtained from the online system of the Ministry of Education. The neural network used was a multilayer perceptron as a backpropagation algorithm optimized through the gradient descent method. To evaluate the obtained results, the mean absolute percentage errors and the Pearson correlation coefficients were calculated. Some hypothesis tests were also carried out in order to verify the existence of significant differences between real values and those obtained by the neural network. The average percentage errors between predicted and actual values varied between 5% and 9%, and the correlation values reached 0,99. The results demonstrated that it is possible to use artificial intelligence as an auxiliary mechanism to plan construction projects, especially in the public sector.

scholarly journals Comparison of neural networks and regression time series when predicting the export development from the USA to PRC

2019 ◽  
Author(s):  
Jakub Horák ◽  
Petr Šuleř ◽  
Jaromír Vrbka
Keyword(s):  
Artificial Intelligence ◽  
Neural Networks ◽  
Time Series ◽  
Artificial Neural Networks ◽  
Linear Regression ◽  
Least Squares ◽  
Weighted Least Squares ◽  
Practice Research ◽  
The Usa ◽  
Artificial Neural

Purpose – artificial neural networks are compared with mixed conclusions in terms of forecasting performance. The most researches indicate that deep-learning models are better than traditional statistical or mathematical models. The purpose of the article is to compare the accuracy of equalizing time series by means of regression analysis and neural networks on the example of the USA export to China. The aim is to show the possible uses and advantages of neural networks in practice. Research methodology – the period for which the data (USA export to the PRC) are available is the monthly balance starting from January 1985 to August 2018. First of all, linear regression as the relatively simple mathematical method is carried out. Subsequently, neural networks as the computational models used in artificial intelligence are used for regression. Findings – in terms of linear regression, the most suitable one appeared to be the curve obtained by means of the least squares methods by negative-exponential smoothing, and the curve obtained by means of the distance-weighted least squares method. In terms of neural networks, all retained structures appeared to be applicable in practice. Artificial neural networks have better representational power than traditional models. Research limitations – the simplification (quite a significant one) appears both in the cases of linear regression and regression by means of neural networks. We work only with two variables – input variable (time) and output variable (USA export to the PRC). Practical implications – in practice, the results – especially the method of artificial neural networks – can be used in the measurement and prediction of the development of exports, but especially in the short term. It can be stated that due to great simplification of the reality it isnʼt possible to predict extraordinary situations and their effect on the USA export to the PRC. Originality/Value – the article focuses on the comparison of two statistical methods, in particular, artificial intelligence is not used in such applications. However, in many economic industries, it has proven better results. It is found that artificial neural networks are able to effectively learn dependencies in and between the time series in the form of export development data.

scholarly journals Aerosol optical depth retrievals at the Izaña Atmospheric Observatory from 1941 to 2013 by using artificial neural networks

2016 ◽  
Vol 9 (1) ◽  
pp. 53-62 ◽  
Author(s):  
R. D. García ◽  
O. E. García ◽  
E. Cuevas ◽  
V. E. Cachorro ◽  
A. Barreto ◽  
...  
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Artificial Neural Networks ◽  
Aerosol Optical Depth ◽  
Optical Depth ◽  
Mineral Dust ◽  
Dust Particles ◽  
Filter Radiometer ◽  
Artificial Neural

Abstract. This paper presents the reconstruction of a 73-year time series of the aerosol optical depth (AOD) at 500 nm at the subtropical high-mountain Izaña Atmospheric Observatory (IZO) located in Tenerife (Canary Islands, Spain). For this purpose, we have combined AOD estimates from artificial neural networks (ANNs) from 1941 to 2001 and AOD measurements directly obtained with a Precision Filter Radiometer (PFR) between 2003 and 2013. The analysis is limited to summer months (July–August–September), when the largest aerosol load is observed at IZO (Saharan mineral dust particles). The ANN AOD time series has been comprehensively validated against coincident AOD measurements performed with a solar spectrometer Mark-I (1984–2009) and AERONET (AErosol RObotic NETwork) CIMEL photometers (2004–2009) at IZO, obtaining a rather good agreement on a daily basis: Pearson coefficient, R, of 0.97 between AERONET and ANN AOD, and 0.93 between Mark-I and ANN AOD estimates. In addition, we have analysed the long-term consistency between ANN AOD time series and long-term meteorological records identifying Saharan mineral dust events at IZO (synoptical observations and local wind records). Both analyses provide consistent results, with correlations  >  85 %. Therefore, we can conclude that the reconstructed AOD time series captures well the AOD variations and dust-laden Saharan air mass outbreaks on short-term and long-term timescales and, thus, it is suitable to be used in climate analysis.

Time Series Forecasting of Cyanobacteria Blooms in the Crestuma Reservoir (Douro River, Portugal) Using Artificial Neural Networks

2006 ◽  
Vol 38 (2) ◽  
pp. 227-237 ◽  
Author(s):  
Luis Oliva Teles ◽  
Vitor Vasconcelos ◽  
Luis Oliva Teles ◽  
Elisa Pereira ◽  
Martin Saker ◽  
...  
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Artificial Neural Networks ◽  
Time Series Forecasting ◽  
Cyanobacteria Blooms ◽  
Artificial Neural ◽  
Douro River

scholarly journals The Application of Artificial Neural Networks in Predicting Children’s Giftedness

2016 ◽  
Vol 19 (1) ◽  
pp. 49-59 ◽  
Author(s):  
Nina Pavlin-Bernardić ◽  
◽  
Silvija Ravić ◽  
Ivan Pavao Matić ◽  
◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
School Readiness ◽  
Multilayer Perceptron ◽  
Gifted Students ◽  
Test Sample ◽  
Hyperbolic Tangent ◽  
Readiness Assessment ◽  
Artificial Neural

Artificial neural networks have a wide use in the prediction and classification of different variables, but their application in the area of educational psychology is still relatively rare. The aim of this study was to examine the accuracy of artificial neural networks in predicting students’ general giftedness. The participants were 221 fourth grade students from one Croatian elementary school. The input variables for artificial neural networks were teachers’ and peers’ nominations, school grades, earlier school readiness assessment and parents’ education. The output variable was the result on the Standard Progressive Matrices (Raven, 1994), according to which students were classified as gifted or non-gifted. We tested two artificial neural networks’ algorithms: multilayer perceptron and radial basis function. Within each algorithm, a number of different types of activation functions were tested. 80% of the sample was used for training the network and the remaining 20% to test the network. For a criterion according to which students were classified as gifted if their result on the Standard Progressive Matrices was in the 95th centile or above, the best model was obtained by the hyperbolic tangent multilayer perceptron, which had a high accuracy of 100% of correctly classified non-gifted students and 75% correctly classified gifted students in the test sample. When the criterion was the 90th centile or above, the best model was also obtained by the hyperbolic tangent multilayer perceptron, but the accuracy was lower: 94.7% in the classification of non-gifted students and 66.7% in the classification of gifted students. The study has shown artificial neural networks’ potential in this area, which should be further explored. Keywords: gifted students, identification of gifted students, artificial neural networks

scholarly journals Use of artificial neural networks and geographic objects for classifying remote sensing imagery

CERNE ◽  
2014 ◽  
Vol 20 (2) ◽  
pp. 267-276 ◽  
Author(s):  
Pedro Resende Silva ◽  
Fausto Weimar Acerbi Júnior ◽  
Luis Marcelo Tavares de Carvalho ◽  
José Roberto Soares Scolforo
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Image Segmentation ◽  
Artificial Neural Networks ◽  
Forest Inventory ◽  
Landsat Tm ◽  
Inventory Data ◽  
Forest Inventory Data ◽  
Artificial Neural ◽  
Test Use

The aim of this study was to develop a methodology for mapping land use and land cover in the northern region of Minas Gerais state, where, in addition to agricultural land, the landscape is dominated by native cerrado, deciduous forests, and extensive areas of vereda. Using forest inventory data, as well as RapidEye, Landsat TM and MODIS imagery, three specific objectives were defined: 1) to test use of image segmentation techniques for an object-based classification encompassing spectral, spatial and temporal information, 2) to test use of high spatial resolution RapidEye imagery combined with Landsat TM time series imagery for capturing the effects of seasonality, and 3) to classify data using Artificial Neural Networks. Using MODIS time series and forest inventory data, time signatures were extracted from the dominant vegetation formations, enabling selection of the best periods of the year to be represented in the classification process. Objects created with the segmentation of RapidEye images, along with the Landsat TM time series images, were classified by ten different Multilayer Perceptron network architectures. Results showed that the methodology in question meets both the purposes of this study and the characteristics of the local plant life. With excellent accuracy values for native classes, the study showed the importance of a well-structured database for classification and the importance of suitable image segmentation to meet specific purposes.

Sign in / Sign up

Export Citation Format

Share Document