Complete ensemble empirical mode decomposition with adaptive noise integrating feedforward neural network for tourist arrival forecasting

Herawati, Sri; Negara, Yudha Dwi Putra; Latif, Muhammad

doi:10.1088/1742-6596/2193/1/012049

Cited by 4 publications

(4 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The residuals of electron densities possess clear wave perturbations. To separate the waves based on their vertical wavelength and time period, an empirical mode decomposition analysis is performed on the residuals of the electron density along the altitude and time, respectively (Herawati et al, 2022;Purba et al, 2018;McDonald et al, 2007).…”

Section: Observations and Data Analysismentioning

confidence: 99%

Multiple Ionospheric Descending Layers Over Arecibo

Dharmalingam

Sivakandan

Sulzer

2023

Preprint

View full text Add to dashboard Cite

Observations using Arecibo Observatory’s highly sensitive Incoherent Scattering Radar (AO-ISR) show ionospheric descending layers from as high as $\sim$400 km, much higher than earlier studies, with continuity down to 90 km. The AO-ISR was operated to observe the ion-line and plasma-line with coded-long-pulse for high temporal and spatial resolution of 35/10 seconds and 300 m, respectively, during 01-06 February 2019. We found multiple layering structures descending from 400 km to 90 km in all these six days. These layers are traditionally called intermediate descending layers (IDLs) ($>$130 km and below F-peak), upper semi-diurnal daytime $\&$ nighttime layers (110 km-130 km), and lower diurnal layers($<$110 km). We have denoted the new daytime descending layers above the hmF2 as top-side descending layers (TDLs). All these layers are collectively named ion descending layers (IonDLs) since all of them are connected with some discontinuity at the F1-peak (i.e., 170 km), except for the daytime lower-diurnal layer. The most pronounced IonDLs occur in the twilight times. IonDLs mainly occur in shear zones of the vertical ion drifts and are favored by downward ion drifts, and their descent speeds increase with increasing altitude. The estimated phase velocities of the waves in the F-region are comparable with the descending speed of the IonDLs. Furthermore, IonDLs/IDLs occur with and without spread-F events but intensified spread-F events raise their beginning altitude. The TDLs and IDLs are driven by gravity waves with time periods of 1.5-4 hours.

show abstract

Section: Observations and Data Analysismentioning

confidence: 99%

Multiple Ionospheric Descending Layers Over Arecibo

Dharmalingam

Sivakandan

Sulzer

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Feed-forward neural network has powerful and diversified functions, which makes it the most popular network. Based on different training methods and different learning styles, feedforward networks have many different variants [14]. Here, two feedforward networks used in this paper are briefly introduced.…”

Section: Feed-forward Neural Networkmentioning

confidence: 99%

Multilayer Performance Improvement of Feedforward Neural Networks

2020

View full text Add to dashboard Cite

The feeder neural network is one of the widely used neural networks. It can transfer input data from the input layer to the output layer without feedback. The feeder neural network is improved by using different techniques and many network models with different functions are obtained. In this thesis, the improvement of the performance of the power supply nervous network is studied and applied. This thesis analyzes two types of neural network feeders, namely neural network and self-coded neural network. Then the algorithm is used to solve the two problems of slow convergence and easy to fall to the local minima in the algorithm of the feeder neural network, and the performance comparison experiment with other depth algorithms is carried out. According to the experimental results, the improved chopped feedforward neural network in this paper is an effective deep learning framework.

show abstract

“…However, the neural network will be in the process of learning because of various reasons: noise data [10], data initialization [11], and other problems, the final training results and the predicted results are far from the results, and the loss is too significant. Therefore, we need to optimize the model constantly.…”

Section: Introductionmentioning

confidence: 99%

Optimize the Performance of the Neural Network by using a Mini Dataset Processing Method

Jingliang Chen,

Chenchen Wu

et al. 2023

Journal of Internet Technology

View full text Add to dashboard Cite

<p>In the case of traditional methods such as network models and algorithms are highly open source and highly bound to hardware, data processing has become an important method to optimize the performance of neural networks. In this paper, we combine traditional data processing methods and propose a method based on the mini dataset which is strictly randomly divided within the training process; and takes the calculation results of the cross-entropy loss function as the measurement standard, by comparing the mini dataset, screening, and processing to optimize the deep neural network. Using this method, each iteration training can obtain a relatively optimal result, and the optimization effects of each time are integrated to optimize the results of each epoch. Finally, in order to verify the effectiveness and applicability of this data processing method, experiments are carried out on MNIST, HAGRID, and CIFAR-10 datasets to compare the effects of using this method and not using this method under different hyper-parameters, and finally, the effectiveness of this data processing method is verified. Finally, we summarize the advantages and limitations of this method and look forward to the future improvement direction of this method.</p> <p> </p>

show abstract

Complete ensemble empirical mode decomposition with adaptive noise integrating feedforward neural network for tourist arrival forecasting

Cited by 4 publications

References 10 publications

Multiple Ionospheric Descending Layers Over Arecibo

Multiple Ionospheric Descending Layers Over Arecibo

Multilayer Performance Improvement of Feedforward Neural Networks

Optimize the Performance of the Neural Network by using a Mini Dataset Processing Method

Contact Info

Product

Resources

About