Brazilian Soil Bulk Density Prediction Based on a Committee of Neural Regressors

Haddad, Diego B.; Assis, Laura Silva de; Tarrataca, Luís; Gomes, Andréa da Silva; Ceddia, Marcos Bacis; Oliveira, Rosane Ferreira de; P., Jurair R. de; Brandão, Diego

doi:10.1109/ijcnn.2018.8489177

Cited by 5 publications

(7 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It may be noted that the accuracy of the system is indeed better than the previous test. For this configuration, it was found that the ANN performs best if the hidden layer has 13 neurons, which confirms the indication of K ‐fold validation (with K = 10) for architecture selection 21 …”

Section: Tests and Resultssupporting

confidence: 75%

“…Validation helps to circumvent the probability of obtaining a biased solution due to the fact that the true risk function is distinct from the empirical risk function, an issue that is termed as overtraining or overfitting 20 . Thus, the ANN learns using the training set, while its performance is evaluated using the validation set for early stopping 21 . The performance of the ANN was then measured using as a parameter the mean squared error (MSE) evaluated on the test dataset.…”

Section: Tests and Resultsmentioning

confidence: 99%

“…20 Thus, the ANN learns using the training set, while its performance is evaluated using the validation set for early stopping. 21 The performance of the ANN was then measured using as a parameter the mean squared error (MSE) evaluated on the test dataset. The entries in the test set are all independent from the training set.…”

Section: Tests and Resultsmentioning

confidence: 99%

“…For this configuration, it was found that the ANN performs best if the hidden layer has 13 neurons, which confirms the indication of K-fold validation (with K = 10) for architecture selection. 21 The last test conducted was an attempt to increase accuracy by reducing even more the spectral spacing between the filters. This test considered seven filters with the following resonant wavelengths: 1505, 1520, 1535, 1550, 1565, 1580, and 1595 nm.…”

Section: Tests and Resultsmentioning

confidence: 99%

See 3 more Smart Citations

Interrogation system for optical sensor using filter bank and artificial neural network

Jucá

Haddad

Santos

2020

Micro & Optical Tech Letters

View full text Add to dashboard Cite

Considering the increasingly wide application of optical fiber sensors, this article aims to present an alternative form of interrogation without the use of an optical spectrum analyzer or any other high-cost devices. The sensor studied here is a long-period grating being used to measure temperature. The interrogator is composed of optical filters and photodetectors whose responses are processed by a suitably trained artificial neural network. Results show that this technique enables effective interrogation of a range limited only by the optical bandwidth of the light source.

show abstract

Section: Tests and Resultssupporting

confidence: 75%

Section: Tests and Resultsmentioning

confidence: 99%

Section: Tests and Resultsmentioning

confidence: 99%

Section: Tests and Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Interrogation system for optical sensor using filter bank and artificial neural network

Jucá

Haddad

Santos

2020

Micro & Optical Tech Letters

View full text Add to dashboard Cite

show abstract

“…The committee is able to combine individual weak predictors in order to produce an improved overall regression [ 34 ]. Given the variation of the data, and considering the reliability issues surrounding multiple data sources, we use the median of the weak predictors to hedge against outliers [ 34 , 35 ]. Figure 11 illustrates the committee strategy.…”

Section: Neural Prediction Of the Brazilian Case Fatality Ratementioning

confidence: 99%

Flattening the curves: on-off lock-down strategies for COVID-19 with an application to Brazil

et al. 2021

View full text Add to dashboard Cite

The current COVID-19 pandemic is affecting different countries in different ways. The assortment of reporting techniques alongside other issues, such as underreporting and budgetary constraints, makes predicting the spread and lethality of the virus a challenging task. This work attempts to gain a better understanding of how COVID-19 will affect one of the least studied countries, namely Brazil. Currently, several Brazilian states are in a state of lock-down. However, there is political pressure for this type of measures to be lifted. This work considers the impact that such a termination would have on how the virus evolves locally. This was done by extending the SEIR model with an on / off strategy. Given the simplicity of SEIR we also attempted to gain more insight by developing a neural regressor. We chose to employ features that current clinical studies have pinpointed has having a connection to the lethality of COVID-19. We discuss how this data can be processed in order to obtain a robust assessment.

show abstract

Toward Low-Cost and Sustainable IoT Systems for Soil Monitoring in Coastal Wetlands

Hao,

Sharma,

Fleming

et al. 2023

2023 IEEE 9th International Conference on Collaboration and Internet Computing (CIC)

View full text Add to dashboard Cite

Brazilian Soil Bulk Density Prediction Based on a Committee of Neural Regressors

Cited by 5 publications

References 45 publications

Interrogation system for optical sensor using filter bank and artificial neural network

Interrogation system for optical sensor using filter bank and artificial neural network

Flattening the curves: on-off lock-down strategies for COVID-19 with an application to Brazil

Toward Low-Cost and Sustainable IoT Systems for Soil Monitoring in Coastal Wetlands

Contact Info

Product

Resources

About