Testing Artificial Neural Network (ANN) for Spatial Interpolation

Nevtipilova, Veronika; Pastwa, Justyna; Boori, Mukesh Singh; Voženílek, Vít

doi:10.4172/2329-6755.1000145

Cited by 15 publications

(14 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…But despite a small difference, due to lower values of MAE and RMSE in neural networks method, this method was chosen as a better method. This result is consistent with that of Sitharam et al (2008) and Nevtipilova et al (2014). They also examined the superiority of ANNs over other methods.…”

Section: Resultssupporting

confidence: 92%

“…In another study, the capability of ANNs in spatial interpolation was evaluated and compared with conventional interpolation methods like Kriging and IDW. The results indicated that despite the high value of RMSE in the MLP method, this method can be used in spatial interpolation (Nevtipilova et al 2014). In an investigation Sitharam et al (2008) compared on geostatistics, neural network and support vector machine methods; the results illustrated a better performance of the ANN method over the other methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Interpolation of soil infiltration in furrow irrigation: Comparison of kriging, inverse distance weighting, multilayer perceptron and principal component analysis methods

Alipour

Nasseri

2019

pjss

View full text Add to dashboard Cite

<p>Study on soil infiltration rate as part of water cycle is essential for managing water resources and designing irrigation systems. The present study was conducted with the aim to compare Kriging, inverse distance weighting (IDW), multilayer perceptron (MLP) and principal component analysis (PCA) methods in the interpolation of soil infiltration in furrow irrigation, and determine the best interpolation method. To conduct infiltration tests, furrows were made on the farm in four triad groups. Infiltration through the blocked furrows method was measured 10, 20, 30, 40, 50, 60, 90, 120, 150, 160, 180 and 210 min after irrigation at a 10-meter distance in each furrow. Data were analyzed by GS+ and Neuro Solutions (NS) software packages. In this study, the maximum error (ME), mean bias error (MBE), mean absolute error (MAE), root mean square error (RMSE), relative error (RE) and correlation coefficient (r) were used to compare the interpolation methods. The results of analysis of variance (ANOVA) indicated that differences in methods based on RMSE, MBE, MAE and ME indices were not significant; however, this difference was significant based on r and RE indices. According to the ANOVA results, it can be said that the PCA method with a r of 0.69 and RE of 31%, was predicted with a higher accuracy as compared to other methods.</p>

show abstract

Section: Resultssupporting

confidence: 92%

Section: Introductionmentioning

confidence: 99%

Interpolation of soil infiltration in furrow irrigation: Comparison of kriging, inverse distance weighting, multilayer perceptron and principal component analysis methods

Alipour

Nasseri

2019

pjss

View full text Add to dashboard Cite

show abstract

“…IDW, RFsp and RFSI had similar performance and were slightly worse than OK. Worse performance of IDW compared to OK in synthetic case studies was also found in Zimmerman et al [50], MacCormak et al [51], and Nevtipilova et al [52]. IDW, RFsp, and RFSI performed differently for different semivariogram nugget-to-sill ratios, ranges, and sample sizes (Figure 4).…”

Section: Rfsi Performancesupporting

confidence: 64%

Random Forest Spatial Interpolation

et al. 2020

View full text Add to dashboard Cite

For many decades, kriging and deterministic interpolation techniques, such as inverse distance weighting and nearest neighbour interpolation, have been the most popular spatial interpolation techniques. Kriging with external drift and regression kriging have become basic techniques that benefit both from spatial autocorrelation and covariate information. More recently, machine learning techniques, such as random forest and gradient boosting, have become increasingly popular and are now often used for spatial interpolation. Some attempts have been made to explicitly take the spatial component into account in machine learning, but so far, none of these approaches have taken the natural route of incorporating the nearest observations and their distances to the prediction location as covariates. In this research, we explored the value of including observations at the nearest locations and their distances from the prediction location by introducing Random Forest Spatial Interpolation (RFSI). We compared RFSI with deterministic interpolation methods, ordinary kriging, regression kriging, Random Forest and Random Forest for spatial prediction (RFsp) in three case studies. The first case study made use of synthetic data, i.e., simulations from normally distributed stationary random fields with a known semivariogram, for which ordinary kriging is known to be optimal. The second and third case studies evaluated the performance of the various interpolation methods using daily precipitation data for the 2016–2018 period in Catalonia, Spain, and mean daily temperature for the year 2008 in Croatia. Results of the synthetic case study showed that RFSI outperformed most simple deterministic interpolation techniques and had similar performance as inverse distance weighting and RFsp. As expected, kriging was the most accurate technique in the synthetic case study. In the precipitation and temperature case studies, RFSI mostly outperformed regression kriging, inverse distance weighting, random forest, and RFsp. Moreover, RFSI was substantially faster than RFsp, particularly when the training dataset was large and high-resolution prediction maps were made.

show abstract

“…The neuralnet package uses supervised learning algorithms for training the network (Günther and Fritsch, 2010). The neuralnet package in the R program was used in many researches (Soni and Abdullahi, 2015;Dengel et al, 2013;Nevtipilova et al, 2014).…”

Section: Development Of Artificial Neural Networkmentioning

confidence: 99%

Evaluation of Interregional Freight Generation Modelling Methods by using Nationwide Commodity Flow Survey Data

Hirun¹

2016

American Journal of Engineering and Applied Sciences

View full text Add to dashboard Cite

A trip generation model is one of the four parts of the classical transport planning model, which explores the volume of trip or freight at the originating and destination points of a traffic analysis zone. The process of calibrating a trip generation model needs appropriate data. Freight transport data are always robust and a powerful calibration technique is required to handle the robustness of such data. The objective of this research is to evaluate the performance of the freight generation model, calibrated by the Artificial Neural Network (ANN), against the conventional linear regression model. The 2012 Thailand commodity flow survey data from National Statistics Organization of Thailand were used for calibration. Interprovincial freight shipment data, across the kingdom of Thailand (77 origins and 77 destinations), were divided into four categories-agricultural products, industrial products, consumer products and construction material. The results indicated that the regression based model failed to accord with the regression assumption, while ANN can also provide the same performance in explaining the relationship between dependent and independent variables. ANN is considered to be a better calibration technique as the concerned data do not accord with regression assumption.

show abstract

Testing Artificial Neural Network (ANN) for Spatial Interpolation

Cited by 15 publications

References 8 publications

Interpolation of soil infiltration in furrow irrigation: Comparison of kriging, inverse distance weighting, multilayer perceptron and principal component analysis methods

Interpolation of soil infiltration in furrow irrigation: Comparison of kriging, inverse distance weighting, multilayer perceptron and principal component analysis methods

Random Forest Spatial Interpolation

Evaluation of Interregional Freight Generation Modelling Methods by using Nationwide Commodity Flow Survey Data

Contact Info

Product

Resources

About