An Improved Pi-Sigma Neural Network with Error Feedback for Physical Time Series Prediction

Akram, Urooj

doi:10.30534/ijatcse/2019/5381.32019

Cited by 9 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The determination of nutritional value (the complex of product properties that provide a person's physiological needs for energy and basic nutrients -proteins, fats, carbohydrates, vitamins, macro -and microelements) and energy value (the amount of energy (kcal, kJ) released in the human body from food substances to ensure its physiological functions) are mandatory in the development of new types of products [12][13][14].…”

Section: Methodsmentioning

confidence: 99%

Analyzing Soft Cheeses With Vegetable Additives With Neural Network

Tamara¹

2020

IJATCSE

View full text Add to dashboard Cite

A wide range of dairy-based products with additives of plant origin is due to the variety of forms of raw materials and methods of processing it. New soft cheeses with vegetable additives have been developed. Physical and chemical parameters of plant raw materials were studied, and it was found that the wheat germ contains (%) -28.42 protein, 6.7 -fat, carbohydrates -19.98. The formula of vegetable additives for the production of soft cheeses has been developed, the ratio of vegetable raw materials and wheat germ is 70:30, while the mass fraction of protein in the additive is 10.6%. Mathematically, neural networks can be considered as a class of statistical modeling methods, which, in turn, can be divided into three classes: probability density estimation, classification and regression. The problem of estimating the probability density by the method of mixing Gaussian distributions is solved with the help of inverse distribution networks and the generalized d-rule.

show abstract

Section: Methodsmentioning

confidence: 99%

Analyzing Soft Cheeses With Vegetable Additives With Neural Network

Tamara¹

2020

IJATCSE

View full text Add to dashboard Cite

show abstract

“…Deep Learning (DL) approaches has been widely used to measure of problems are included computer vision, natural language processing [10], automatic time series forecasting [11], hand-writing recognition [12], and financial time series prediction [13] to name a few, outperformed existing baseline scheme. However, some DL approaches were used financial time series data, and have been applied many classification tasks are included, text-based classification, portfolio optimization, volatility predicting and price base prediction.…”

Section: Related Workmentioning

confidence: 99%

“…We replace hyperbolic tangent activation function (tanh) with (Relu) in equation (11) the new equation is: (13) Update gate z t updates the following information sending by previous activation with time step t and pass to the next step. On the other hand, reset gate r t control the short-term dependencies and long-term dependencies has controlled by update gates z t in the GRU model.…”

Section: Gru-cnn Proposed Architecturementioning

confidence: 99%

Predicting Financial Prices of Stock Market using Recurrent Convolutional Neural Networks

Zulqarnain¹,

Ghazali²,

Ghouse³

et al. 2020

IJISA

View full text Add to dashboard Cite

Financial time-series prediction has been long and the most challenging issues in financial market analysis. The deep neural networks is one of the excellent data mining approach has received great attention by researchers in several areas of time-series prediction since last 10 years. “Convolutional neural network (CNN) and recurrent neural network (RNN) models have become the mainstream methods for financial predictions. In this paper, we proposed to combine architectures, which exploit the advantages of CNN and RNN simultaneously, for the prediction of trading signals. Our model is essentially presented to financial time series predicting signals through a CNN layer, and directly fed into a gated recurrent unit (GRU) layer to capture long-term signals dependencies. GRU model perform better in sequential learning tasks and solve the vanishing gradients and exploding issue in standard RNNs. We evaluate our model on three datasets for stock indexes of the Hang Seng Indexes (HSI), the Deutscher Aktienindex (DAX) and the S&P 500 Index range 2008 to 2016, and associate the GRU-CNN based approaches with the existing deep learning models. Experimental results present that the proposed GRU-CNN model obtained the best prediction accuracy 56.2% on HIS dataset, 56.1% on DAX dataset and 56.3% on S&P500 dataset respectively.

show abstract

“…The stability is depending on the types of data and factors affecting them [45,46]. For instance, the time series signals were observed on a highly non-stationary and/or non-linear range [47,48]. Non-stationary is a common property to vary time-series models, which means, a variable has no clear tendency to return to a constant value or a linear trend.…”

Section: A Data Preparationmentioning

confidence: 99%

A Modified Weight Optimization for Artificial Higher Order Neural Networks in Physical Time Series

Husaini¹,

Ghazali²,

Arbaiy³

et al. 2020

IJACSA

View full text Add to dashboard Cite

Many methods and approaches have been proposed for analyzing and forecasting time series data. There are different Neural Network (NN) variations for specific tasks (e.g., Deep Learning, Recurrent Neural Networks, etc.). Time series forecasting are a crucial component of many important applications, from stock markets to energy load forecasts. Recently, Swarm Intelligence (SI) techniques including Cuckoo Search (CS) have been established as one of the most practical approaches in optimizing parameters for time series forecasting. Several modifications to the CS have been made, including Modified Cuckoo Search (MCS) that adjusts the parameters of the current CS, to improve algorithmic convergence rates. Therefore, motivated by the advantages of these MCSs, we use the enhanced MCS known as the Modified Cuckoo Search-Markov Chain Monté Carlo (MCS-MCMC) learning algorithm for weight optimization in Higher Order Neural Networks (HONN) models. The Lévy flight function in the MCS is replaced with Markov Chain Monté Carlo (MCMC) since it can reduce the complexity in generating the objective function. In order to prove that the MCS-MCMC is suitable for forecasting, its

show abstract

An Improved Pi-Sigma Neural Network with Error Feedback for Physical Time Series Prediction

Cited by 9 publications

References 0 publications

Analyzing Soft Cheeses With Vegetable Additives With Neural Network

Analyzing Soft Cheeses With Vegetable Additives With Neural Network

Predicting Financial Prices of Stock Market using Recurrent Convolutional Neural Networks

A Modified Weight Optimization for Artificial Higher Order Neural Networks in Physical Time Series

Contact Info

Product

Resources

About