A new formulation for feedforward neural networks

Artificial neural network-based shelf life prediction approach in the food storage process: A review

The prediction of food shelf life has become a vital tool for distributors and consumers, enabling them to determine storage and optimal edible time, thus avoiding unexpected food waste. Artificial neural network (ANN) have emerged as an effective, fast and accurate method for modeling, simulating and predicting shelf life in food. ANNs are capable of tackling nonlinear, complex and ill-defined problems between the variables without prior knowledge. ANN model exhibited excellent fit performance evidenced by low root mean squared error and high correlation coefficient. The low relative error between actual values and predicted values from the ANN model demonstrates its high accuracy. This paper describes the modeling of ANN in food quality prediction, encompassing commonly used ANN architectures, ANN simulation techniques, and criteria for evaluating ANN model performance. The review focuses on the application of ANN for modeling nonlinear food quality during storage, including dairy, meat, aquatic, fruits, and vegetables products. The future prospects of ANN development mainly focus on optimal models and learning algorithm selection, multiple model fusion, self-learning and self-correcting shelf-life prediction model development, and the potential utilization of deep learning techniques.

Training of Feed-Forward Neural Networks by Using Optimization Algorithms Based on Swarm-Intelligent for Maximum Power Point Tracking

One of the most used artificial intelligence techniques for maximum power point tracking is artificial neural networks. In order to achieve successful results in maximum power point tracking, the training process of artificial neural networks is important. Metaheuristic algorithms are used extensively in the literature for neural network training. An important group of metaheuristic algorithms is swarm-intelligent-based optimization algorithms. In this study, feed-forward neural network training is carried out for maximum power point tracking by using 13 swarm-intelligent-based optimization algorithms. These algorithms are artificial bee colony, butterfly optimization, cuckoo search, chicken swarm optimization, dragonfly algorithm, firefly algorithm, grasshopper optimization algorithm, krill herd algorithm, particle swarm optimization, salp swarm algorithm, selfish herd optimizer, tunicate swarm algorithm, and tuna swarm optimization. Mean squared error is used as the error metric, and the performances of the algorithms in different network structures are evaluated. Considering the results, a success ranking score is obtained for each algorithm. The three most successful algorithms in both training and testing processes are the firefly algorithm, selfish herd optimizer, and grasshopper optimization algorithm, respectively. The training error values obtained with these algorithms are 4.5 × 10-4, 1.6 × 10-3, and 2.3 × 10-3, respectively. The test error values are 4.6 × 10-4, 1.6 × 10-3, and 2.4 × 10-3, respectively. With these algorithms, effective results have been achieved in a low number of evaluations. In addition to these three algorithms, other algorithms have also achieved mostly acceptable results. This shows that the related algorithms are generally successful ANFIS training algorithms for maximum power point tracking.

Numerical Simulation for the Sound Absorption Properties of Ceramic Resonators

This work reports the results of experimental measurements of the sound absorption coefficient of ceramic materials using the principle of acoustic resonators. Subsequently, the values obtained from the measurements were used to train a simulation model of the acoustic behavior of the analyzed material based on artificial neural networks. The possible applications of sound-absorbing materials made with ceramic can derive from aesthetic or architectural needs or from functional needs, as ceramic is a fireproof material resistant to high temperatures. The results returned by the simulation model based on the artificial neural networks algorithm are particularly significant. This result suggests the adoption of this technology to find the finest possible configuration that allows the best sound absorption performance of the material.

Transboundary Basins Need More Attention: Anthropogenic Impacts on Land Cover Changes in Aras River Basin, Monitoring and Prediction

Changes in land cover (LC) can alter the basin hydrology by affecting the evaporation, infiltration, and surface and subsurface flow processes, and ultimately affect river water quantity and quality. This study aimed to monitor and predict the LC composition of a major, transboundary basin contributing to the Caspian Sea, the Aras River Basin (ARB). To this end, four LC maps of ARB corresponding to the years 1984, 2000, 2010, and 2017 were generated using Landsat satellite imagery from Armenia and the Nakhchivan Autonomous Republic. The LC gains and losses, net changes, exchanges, and the spatial trend of changes over 33 years (1984–2017) were investigated. The most important drivers of these changes and the most accurate LC transformation scenarios were identified, and a land change modeler (LCM) was applied to predict the LC change for the years 2027 and 2037. Validation results showed that LCM, with a Kappa index higher than 81%, is appropriate for predicting LC changes in the study area. The LC changes observed in the past indicate significant anthropogenic impacts on the basin, mainly by constructing new reservoir dams and expanding agriculture and urban areas, which are the major water-consuming sectors. Results show that over the past 33 years, agricultural areas have grown by more than 57% from 1984 to 2017 in the study area. Results also indicate that the given similar anthropogenic activities will keep on continuing in the ARB, and agricultural areas will increase by 2% from 2017 to 2027, and by another 1% from 2027 to 2037. Results of this study can support transboundary decision-making processes to analyze potential adverse impacts following past policies with neighboring countries that share the same water resources.

Biogas production estimation using data-driven approaches for cold region municipal wastewater anaerobic digestion

The objective of this study was to estimate biogas (including methane, carbon dioxide and hydrogen sulphide) production rates from the anaerobic digesters at the Saskatoon Wastewater Treatment Plant (SWTP), Saskatchewan, Canada. Average daily ambient temperatures typically fluctuate between -40 °C and 30 °C over the year making the management of the SWTP processes challenging. Operating parameters were taken from 2014 to 2016 including volatile fatty acids (VFAs), total solids, fixed solids, volatile solids, pH, and inflow rate. The input parameters were processed using two methods including a correlation test and principal component analysis (PCA) to determine highly correlated variables prior to use in models. The two models used to estimate biogas production rates are a multi-layered perceptron feed forward artificial neural network (ANN) and an adaptive network-based fuzzy inference system (ANFIS) with grid partition (GP), subtractive clustering (SC) and fuzzy c-means clustering (FCMC). The models using PCA processed variables had reasonable performances with shorter model processing times, while reducing model input data. Among various structures of ANN and ANFIS models for estimation of biogas generation, the ANFIS-FCMC results had better agreement with the observed data. Its average approximation of emission rates of CH₄, CO₂ and H₂S from the wastewater digesters were 3,086, 6,351, and 41.5 g/min, respectively. Our group is assessing similar estimation methodology for the remaining SWTP wastewater treatment processes that are more highly impacted by the seasonal temperature variations including primary and secondary treatment processes.

A Comparison of Regularization Techniques in Deep Neural Networks

Artificial neural networks (ANN) have attracted significant attention from researchers because many complex problems can be solved by training them. If enough data are provided during the training process, ANNs are capable of achieving good performance results. However, if training data are not enough, the predefined neural network model suffers from overfitting and underfitting problems. To solve these problems, several regularization techniques have been devised and widely applied to applications and data analysis. However, it is difficult for developers to choose the most suitable scheme for a developing application because there is no information regarding the performance of each scheme. This paper describes comparative research on regularization techniques by evaluating the training and validation errors in a deep neural network model, using a weather dataset. For comparisons, each algorithm was implemented using a recent neural network library of TensorFlow. The experiment results showed that an autoencoder had the worst performance among schemes. When the prediction accuracy was compared, data augmentation and the batch normalization scheme showed better performance than the others.

Biologically inspired intelligent decision making: a commentary on the use of artificial neural networks in bioinformatics

Artificial neural networks (ANNs) are a class of powerful machine learning models for classification and function approximation which have analogs in nature. An ANN learns to map stimuli to responses through repeated evaluation of exemplars of the mapping. This learning approach results in networks which are recognized for their noise tolerance and ability to generalize meaningful responses for novel stimuli. It is these properties of ANNs which make them appealing for applications to bioinformatics problems where interpretation of data may not always be obvious, and where the domain knowledge required for deductive techniques is incomplete or can cause a combinatorial explosion of rules. In this paper, we provide an introduction to artificial neural network theory and review some interesting recent applications to bioinformatics problems.

MODIFIED FUZZY NEURAL NETWORK FOR THE CLASSIFICATION OF MURDER CASES IN CRIMINAL LAW USING GAUSSIAN MEMBERSHIP FUNCTION

This paper presents the problem of decision making by a judge in the case of murder cases in criminal law using single hidden layered fuzzy neural network algorithm. Since the membership functions (MFs) of fuzzy sets can affect the performance of the classification models, determination of MFs is crucial. In this paper, the MF selected is Triangular and Gaussian is proposed for evaluation to improve the classification results. To evaluate the effectiveness of the proposed Fuzzy Neural Network model for the classification of murder cases, sufficient number of real-world data sets of court decisions are trained and tested. The simulation model of different membership functions for the modified fuzzy neural network architecture is implemented in C++. Experimental results show that the proposed neuro-fuzzy classifier with Gaussian MF outperforms Triangular MF with higher accuracy. The proposed classification model is proved to be a suitable tool for classification of murder cases in criminal law.

Common nature of learning between back-propagation and Hopfield-type neural networks for generalized matrix inversion with simplified models

In this paper, two simple-structure neural networks based on the error back-propagation (BP) algorithm (i.e., BP-type neural networks, BPNNs) are proposed, developed, and investigated for online generalized matrix inversion. Specifically, the BPNN-L and BPNN-R models are proposed and investigated for the left and right generalized matrix inversion, respectively. In addition, for the same problem-solving task, two discrete-time Hopfield-type neural networks (HNNs) are developed and investigated in this paper. Similar to the classification of the presented BPNN-L and BPNN-R models, the presented HNN-L and HNN-R models correspond to the left and right generalized matrix inversion, respectively. Comparing the BPNN weight-updating formula with the HNN state-transition equation for the specific (i.e., left or right) generalized matrix inversion, we show that such two derived learning-expressions turn out to be the same (in mathematics), although the BP and Hopfield-type neural networks are evidently different from each other a great deal, in terms of network architecture, physical meaning, and training patterns. Numerical results with different illustrative examples further demonstrate the efficacy of the presented BPNNs and HNNs for online generalized matrix inversion and, more importantly, their common natures of learning.

Enhancing weak signal transmission through a feedforward network

The ability to transmit and amplify weak signals is fundamental to signal processing of artificial devices in engineering. Using a multilayer feedforward network of coupled double-well oscillators as well as Fitzhugh-Nagumo oscillators, we here investigate the conditions under which a weak signal received by the first layer can be transmitted through the network with or without amplitude attenuation. We find that the coupling strength and the nodes' states of the first layer act as two-state switches, which determine whether the transmission is significantly enhanced or exponentially decreased. We hope this finding is useful for designing artificial signal amplifiers.

Performance evaluation of multilayer perceptrons for discriminating and quantifying multiple kinds of odors with an electronic nose

This paper studies several types and arrangements of perceptron modules to discriminate and quantify multiple odors with an electronic nose. We evaluate the following types of multilayer perceptron. (A) A single multi-output (SMO) perceptron both for discrimination and for quantification. (B) An SMO perceptron for discrimination followed by multiple multi-output (MMO) perceptrons for quantification. (C) An SMO perceptron for discrimination followed by multiple single-output (MSO) perceptrons for quantification. (D) MSO perceptrons for discrimination followed by MSO perceptrons for quantification, called the MSO-MSO perceptron model, under the following conditions: (D1) using a simple one-against-all (OAA) decomposition method; (D2) adopting a simple OAA decomposition method and virtual balance step; and (D3) employing a local OAA decomposition method, virtual balance step and local generalization strategy all together. The experimental results for 12 kinds of volatile organic compounds at 85 concentration levels in the training set and 155 concentration levels in the test set show that the MSO-MSO perceptron model with the D3 learning procedure is the most effective of those tested for discrimination and quantification of many kinds of odors.