Cascade-forward Back Propagation Neural Network Research Articles

Analyzing brain-activation responses to auditory stimuli improves the diagnosis of a disorder of consciousness by non-linear dynamic analysis of the EEG

Although auditory stimuli benefit patients with disorders of consciousness (DOC), the optimal stimulus remains unclear. We explored the most effective electroencephalography (EEG)-tracking method for eliciting brain responses to auditory stimuli and assessed its potential as a neural marker to improve DOC diagnosis. We collected 58 EEG recordings from patients with DOC to evaluate the classification model’s performance and optimal auditory stimulus. Using non-linear dynamic analysis (approximate entropy [ApEn]), we assessed EEG responses to various auditory stimuli (resting state, preferred music, subject’s own name [SON], and familiar music) in 40 patients. The diagnostic performance of the optimal stimulus-induced EEG classification for vegetative state (VS)/unresponsive wakefulness syndrome (UWS) and minimally conscious state (MCS) was compared with the Coma Recovery Scale-Revision in 18 patients using the machine learning cascade forward backpropagation neural network model. Regardless of patient status, preferred music significantly activated the cerebral cortex. Patients in MCS showed increased activity in the prefrontal pole and central, occipital, and temporal cortices, whereas those in VS/UWS showed activity in the prefrontal and anterior temporal lobes. Patients in VS/UWS exhibited the lowest preferred music-induced ApEn differences in the central, middle, and posterior temporal lobes compared with those in MCS. The resting state ApEn value of the prefrontal pole (0.77) distinguished VS/UWS from MCS with 61.11% accuracy. The cascade forward backpropagation neural network tested for ApEn values in the resting state and preferred music-induced ApEn differences achieved an average of 83.33% accuracy in distinguishing VS/UWS from MCS (based on K-fold cross-validation). EEG non-linear analysis quantifies cortical responses in patients with DOC, with preferred music inducing more intense EEG responses than SON and familiar music. Machine learning algorithms combined with auditory stimuli showed strong potential for improving DOC diagnosis. Future studies should explore the optimal multimodal sensory stimuli tailored for individual patients.Trial registration: The study is registered in the Chinese Registry of Clinical Trials (Approval no: KYLL-2023-414, Registration code: ChiCTR2300079310).

Boosting flood routing prediction performance through a hybrid approach using empirical mode decomposition and neural networks: a case study of the Mera River in Ankara

Abstract Flood routing is vital in helping to reduce the impact of floods on people and communities by allowing timely and appropriate responses. In this study, the empirical mode decomposition (EMD) signal decomposition technique is combined with cascade forward backpropagation neural network (CFBNN) and feed-forward backpropagation neural network (FFBNN) machine learning (ML) techniques to model 2014 floods in Ankara, Mera River. The data are split in order to avoid the underfitting and overfitting problems of the algorithm. While establishing the algorithm, 70% of the data were divided into training, 15% testing and 15% validation. Graphical indicators and statistical parameters were used for the analysis of model performance. As a result, the EMD signal decomposition technique has been found to improve the performance of ML models. In addition, the EMD-FFBNN hybrid model showed the most accurate estimation results in the flood routing calculation. The study's outputs can assist in designing flood control structures such as levees and dams to help reduce flood risk.

Improved Feed-Forward Backpropagation Neural Network Based on Marine Predators Algorithm for Tuning Automatic Voltage Regulator

This research will discuss the application of an automatic voltage regulator based on the feed-forward back propagation neural network (FFBNN), which is enhanced by the marine predator algorithm (MPA). The marine predators algorithm is a method that adopts marine ecosystem life that is identified in the relationship between predators and prey. MPA is adopting a natural approach to arranging the best food search strategies and finding the latest strategy. The focus of the research is on the performance of speed and rotor angle. The performance of the proposed method will be tested using hidden layer variations. In addition, the proposed method will be compared with the feed-forward backpropagation neural network (FFBNN), cascade-forward backpropagation neural network (CFBNN), Elman recurrent neural network (E-RNN), and Focused Time Delay neural network (FTDNN). The speed and rotor angle of the proposed method have good values. The MPA-FFBNN results are not much different from other methods. The experimental results show that the performance of the proposed method has promising results.

PIGNUS: A Deep Learning model for IDS in industrial internet-of-things

The heterogeneous nature of the Industrial Internet of Thing (IIoT) has a considerable impact on the development of an effective Intrusion Detection System (IDS). The proliferation of linked devices results in multiple inputs from industrial sensors. IDS faces challenges in analyzing the features of the traffic and identifying anonymous behavior. Due to the unavailability of a comprehensive feature mapping method, the present IDS solutions are non-usable to identify zero-day vulnerabilities.In this paper, we introduce the first comprehensive IDS framework that combines an efficient feature-mapping technique and cascading model to solve the above-mentioned problems. We call our proposed solution deeP learnIG model intrusioN detection in indUStrial internet-of things (PIGNUS). PIGNUS integrates Auto Encoders (AE) to select optimal features and Cascade Forward Back Propagation Neural Network (CFBPNN) for classification and attack detection. The cascading model uses interconnected links from the initial layer to the output layer and determines the normal and abnormal behavior patterns and produces a perfect classification. We execute a set of experiments on five popular IIoT datasets: gas pipeline, water storage tank, NSLKDD+, UNSW-NB15, and X-IIoTID. We compare PIGNUS to the state-of-the-art models in terms of accuracy, False Positive Ratio (FPR), precision, and recall. The results show that PIGNUS provides more than 95% accuracy, which is 25% better on average than the existing models. In the other parameters, PIGNUS shows 20% improved FPR, 10% better recall, and 10% better in precision. Overall, PIGNUS proves its efficiency as an IDS solution for IIoTs. Thus, PIGNUS is an efficient solution for IIoTs.

Machine Learning and Statistical Techniques for Daily Wind Energy Prediction

This paper presents the development of wind energy prediction models for the Nala Danavi wind farm in Sri Lanka by using machine learning and statistical techniques. Wind speed and ambient temperature were used as the input variables in modeling while the daily wind energy production was the output variable. Correlation between the wind energy and each weather index was investigated using the Pearson’s and Spearman’s correlation coefficients and it was found that daily wind energy output is positively correlated with both daily averaged input variables. Statistical prediction models of Multiple Linear Regression (MLR) and Power Regression (PR) and the machine learning techniques of Support Vector Regression (SVR), Gaussian Process Regression (GPR), Feed Forward Backpropagation Neural Network (FFBPNN), Cascade-Forward Backpropagation Neural Network (CFBPNN) and Recurrent Neural Network (RNN) were developed. The accuracy of the prediction models was measured in terms of the coefficient of determination, Bias, Percent Root mean square error (RMSE)Bias, and Nash-Sutcliffe Efficiency (NSE). Results of the performance evaluation indicated that all the models are highly accurate while the FFBPNN-based model demonstrates outstanding performance with very low error. Such prediction models are highly important for a country like Sri Lanka whose power generation mainly depends on imported coal followed by hydropower and expanding the on-shore and off-shore wind farms gradually in many potential locations scattered over the country.

Artificial neural network model of catalytic coal gasification in fixed bed

Catalytic coal gasification is a cost-effective way to utilize coal to produce hydrogen or methane. Because of complex interactions among hydrodynamics and chemical reactions with catalyst, it was very difficult to predict hydrogen yield and carbon conversion. Therefore, three kinds of artificial neural network (ANN) models, including feed-forward back propagation neural network (FFBP), cascade-forward back propagation neural network with Levenberge Marquardt algorithms (CFBP), and cascade-forward back propagation neural network with genetic algorithm (CFBP-GA) were used to predict these processes. Three kinds of input parameters were used, such as coal ultimate analyses, coal proximate analyses and operation conditions. Gas yield and carbon conversion were taken as output parameters. R2 of all three established ANN models were above 0.9. The CFBP-GA showed good performance with E2 = 0.000241 and R2 = 0.9978 than the other two ANN models. And then, the model was used to predict the effects of temperature, catalyst type and catalyst loading on carbon conversion and hydrogen yield. The most important three factors for carbon conversion and hydrogen yield were catalyst type, catalyst loading and temperature based on relative importance analysis. The good performances were indicated the ANN model was an effective way to predict hydrogen production and carbon conversion of catalytic coal gasification in fixed bed quickly and accurately, and it could be also extended to other carbon resources to produce hydrogen.

DeBot: A deep learning-based model for bot detection in industrial internet-of-things

In this paper, we show a deep learning model for bot detection, named as DeBot, for industrial network traffic. DeBot uses a novel Cascade Forward Back Propagation Neural Network (CFBPNN) model with a subset of features using the Correlation-based Feature Selection (CFS) technique. A time series-based Nonlinear Auto-regressive Network with eXogenous inputs (NARX) technique analyzes the factors having a higher impact on the target variable and predict the behavioral pattern. To the best of our knowledge, we pioneer the use of optimal feature selection and integration with the cascading model of deep learning in bot detection of IIoTs. We conduct a thorough set of experiments on five popular bot datasets: NF-UNSW-NB15, NF-ToN-IoT, NF-BoT-IoT, NF-CSE-CIC-IDS2018, and ToN-IoT-Windows. We compare CFBPNN with other existing neural network models. We observe that CFBPNN in DeBot shows 100% accuracy in all the datasets with subset evaluation and obtains optimum F1-score and zero precision.

Amphibian species detection in water reservoirs using artificial neural networks for ecology-friendly city planning

Planning cities regardless of natural resources has led to the existence of threats for several species that occupy an important place in local ecosystems. Constructing buildings, bridges, or motorways over water reservoirs is perilous for resident amphibians, which belong to one of the most fragile animal groups. An automated system that reliably identifies whether any amphibian species live in a water reservoir before construction in that region will help building inspectors during city planning and will ensure that citizens can enjoy the opportunities provided by nature. In this paper, an intelligent amphibian species detection system is developed using features extracted from GIS and satellite images. The presence of seven amphibian species is detected using an artificial neural network (ANN). Depending on the spatial features of the region, the proposed cascade-forward backpropagation neural network (CFBNN) model, a special type of ANN, determines which amphibian species live there to enable ecological precautions to be taken before construction starts and to preserve the biodiversity for the future health of the community. The results clearly demonstrate that the proposed approach significantly outperforms recently suggested strategies with more than 15% improvement on average. The performance of the system shows that the proposed detection approach can be a useful tool for smart and ecological city planning.

Robust Process Parameter Design Methodology: A New Estimation Approach by Using Feed-Forward Neural Network Structures and Machine Learning Algorithms

In robust design (RD) modeling, the response surface methodology (RSM) based on the least-squares method (LSM) is a useful statistical tool for estimating functional relationships between input factors and their associated output responses. Neural network (NN)-based models provide an alternative means of executing input-output functions without the assumptions necessary with LSM-based RSM. However, current NN-based estimation methods do not always provide suitable response functions. Thus, there is room for improvement in the realm of RD modeling. In this study, a new NN-based RD modeling procedure is proposed to obtain the process mean and standard deviation response functions. Second, RD modeling methods based on the feed-forward back-propagation neural network (FFNN), cascade-forward back-propagation neural network (CFNN), and radial basis function network (RBFN) are proposed. Third, two simulation studies are conducted using a given true function to verify the proposed three methods. Fourth, a case study is examined to illustrate the potential of the proposed approach. In conclusion, a comparative analysis of the three feed-forward NN structure-based modeling methods and conventional LSM-based RSM proposed in this study showed that the proposed methods were significantly lower in the expected quality loss (EQL) and various variability indicators.

Designing of a rigorous image retrieval system with amalgamation of artificial intelligent techniques and relevance feedback

Retrieving out the most comparable images from huge databases is the challenging task for image retrieval systems. So, there is a great need of constructing a capable and rigorous image retrieval system. In this implementation, an exclusive and competent Content based image retrieval (CBIR) system is schemed by the integration of Color moment (CM) and Local binary pattern (LBP). A hybrid feature vector is created by the combination of these two techniques through the process of normalization. This hybrid feature vector is given as the input to the intelligent classifiers i.e. Support vector machine (SVM) and Cascade forward back propagation neural network (CFBPNN). After that, Relevance feedback (RF) technique is applied so as to get the high level information in order to reduce the semantic gap. So, here two Artificial Intelligent CBIR models are proposed, first one is (Hybrid+SVM+RF) and second is (Hybrid+CFBPNN+RF) and their performance parameters are compared. The implementations are performed on two benchmark dataset Corel-1K and Oxford flower dataset which contains 1000 and 1360 images respectively. Different parameters are figured such as accuracy, precision, average retrieval time, recall etc. The average precision obtained for the first model is 93% with Corel 1K database and 91% with Oxford flower database. And similarly for the second model, it is 97% and 94% respectively which is higher than the first model. This implemented technique is validated on both the datasets and the attained results outperforms with other related s approaches.

Review on the Development of Mining Method Selection to Identify New Techniques Using a Cascade‐Forward Backpropagation Neural Network

The most crucial event in a mining project is the selection of an appropriate mining method (MMS). Consequently, determining the optimal choice is critical because it impacts most of the other key decisions. This study provides a concise overview of the development of multiple selection methods using a cascade‐forward backpropagation neural network (CFBPNN). Numerous methods of multicriteria decision‐making (MCDM) are discussed and compared herein. The comparison includes several factors, such as applicability, subjectivity, qualitative and quantitative data, sensitivity, and validity. The application of artificial intelligence is presented and discussed using CFBPNN. The Chengchao iron mine was selected for this investigation to pick the optimum mining method. The results revealed that cut and fill stoping is the most appropriate mining method, followed by sublevel and shrinkage stoping methods. The least appropriate method is open‐pit mining, followed by room and pillar and longwall mining methods.

Sensitivity Identification of Low-frequency Cantilever Fibre Bragg Grating Accelerometer

Vibration response of low-frequency cantilever fibre Bragg grating (FBG) accelerometer produced by Euler–Bernoulli model (namely FBG-MM model) is found to be frequency-dependent, unsimilar to SDOF model. Therefore, the sensitivity of the cantilever FBG accelerometer could not be identified using polynomial or basic fitting methods. This paper presents the use of cascade-forward backpropagation neural network (CFB) to predict the sensitivity of the cantilever FBG accelerometer in a "black box", which refers to the behaviour of the deep neuralnetwork. The inputs of the network are maximum base accelerations and forcing frequencies, which was set between 20 and 90 Hz (below than the first fundamental frequency of the proposed FBG accelerometer), while the output is the wavelength shift. The validation results show that the wavelength shift predicted by the trained CFB demonstrates good agreement with the FBG-MM, with the input parameter within the range of that used in training process. In addition, results also show that the trained CFB would be invalid if the input parameter is out of the range of that used in training process. In real acceleration measurement, since the forcing frequency is unknown beforehand, the trained CFB must be re-trained by considering the maximum base accelerations are embedded with forcing frequencies.

An Anatomy of a Hybrid Color Descriptor with a Neural Network Model to Enhance the Retrieval Accuracy of an Image Retrieval System

Background: In order to retrieve a particular image from vast repository of images, an efficient system is required and such an eminent system is well-known by the name Content-Based Image Retrieval (CBIR) system. Color is indeed an important attribute of an image and the proposed system consist of a hybrid color descriptor which is used for color feature extraction. Deep learning has gained a prominent importance in the current era. So, the performance of this fusion based color descriptor is also analyzed in the presence of Deep learning classifiers. Method: This paper describes a comparative experimental analysis on various color descriptors and the best two are chosen to form an efficient color based hybrid system denoted as combined color moment-color autocorrelogram (Co-CMCAC). Then, to increase the retrieval accuracy of the hybrid system, a Cascade forward back propagation neural network (CFBPNN) is used. The classification accuracy obtained by using CFBPNN is also compared to Patternnet neural network. Results: The results of the hybrid color descriptor depict that the proposed system has superior results in the order 95.4%, 88.2%, 84.4% and 96.05% for Corel-1K, Corel-5K, Corel-10K and Oxford flower benchmark datasets respectively as compared to many state-of-the-art related techniques. Conclusion: This paper depicts an experimental and analytical analysis on different color feature descriptors namely, Color Moment (CM), Color Auto-Correlogram (CAC), Color Histogram (CH), Color coherence vector (CCV) and Dominant Color Descriptor (DCD). The proposed hybrid Color Descriptor (Co-CMCAC) is utilized for the withdrawal of color features with Cascade Forward Back Propagation Neural Network (CFBPNN) used as a classifier on four benchmark datasets namely Corel-1K, Corel-5K and Corel-10K and Oxford flower.

VOC removal in rotating packed bed: ANN model vs empirical model

The discharged volatile organic compounds (VOCs) have aroused more and more attention because of increasing serious air pollution. Accurate prediction of the emission is very important for the industry. Therefore, two kinds of model were developed to predict the VOCs removal efficiency in the RPB, including empirical model and artificial neural network (ANN) models. The empirical model was mainly including gas and liquid mass transfer, effective contact area and liquid holdup. The ANN models were including Cascade-forward back propagation neural network, Feed-forward distributed time delay neural network, Feed-forward back propagation neural network and Elman-forward back propagation neural network. The input parameters were dimensionless numbers, such as high gravity factor, liquid Reynolds number, gas Reynolds number and dimensionless Henry coefficient. The output parameter was removal efficiency. The outlet concentrations of different VOCs predicted by the ANN model were much better than those of the empirical model. And the disadvantages and the advantages were also analyzed. The effects of high gravity factor, gas flow rate and liquid flow rate on the removal efficiency in the RPB were simulated by ANN model with high precision. And then, operation conditions could be timely adjusted to meet the emission standards.

Improving Neural Network Based on Seagull Optimization Algorithm for Controlling DC Motor

This article presents a direct current (DC) motor control approach using a hybrid Seagull Optimization Algorithm (SOA) and Neural Network (NN) method. SOA method is a nature-inspired algorithm. DC motor speed control is very important to maintain the stability of motor operation. The SOA method is an algorithm that duplicates the life of the seagull in nature. Neural network algorithms will be improved using the SOA method. The neural network used in this study is a feed-forward neural network (FFNN). This research will focus on controlling DC motor speed. The efficacy of the proposed method is compared with the Proportional Integral Derivative (PID) method, the Feed Forward Neural Network (FFNN), and the Cascade Forward Backpropagation Neural Network (CFBNN). From the results of the study, the proposed control method has good capabilities compared to standard neural methods, namely FFNN and CFBNN. Integral Time Absolute Error and Square Error (ITAE and ITSE) values from the proposed method are on average of 0.96% and 0.2% better than the FFNN and CFBNN methods.

An Improved Neural Network Based on Parasitism – Predation Algorithm for an Automatic Voltage Regulator

The parasitism – predation algorithm (PPA) is an optimization method that duplicates the interaction of mutualism between predators (cats), parasites (cuckoos), and hosts (crows). The study employs a combination of the PPA methods using the cascade-forward backpropagation neural network. This hybrid method employs an automatic voltage regulator (AVR) on a single machine system, with the performance measurement focusing on speed and the rotor angle. The performance of the proposed method is compared with the feed-forward backpropagation neural network (FFBNN), cascade-forward backpropagation neural network (CFBNN), Elman recurrent neural network (E-RNN), focused time-delay neural network (FTDNN), and distributed time-delay neural network (DTDNN). The results show that the proposed method exhibits the best speed and rotor angle performance. The PPA-CFBNN method has the ability to reduce the overshoot of the speed by 1.569% and the rotor angle by 0.724%.

Development of surrogate predictive models for the nonlinear elasto-plastic response of medium density fibreboard-based sandwich structures

Medium-density fibreboard (MDF) belongs to a class of engineered wood products facilitating efficient use of wood wastes. For this class of materials, the development of predictive models is crucial for the simulation of their responses under mechanical loads. In this study, samples of sandwich structures based on MDF as the skins and a mushroom-based foam as the core are fabricated and tested under edgewise compression tests. Results from the tests support the idea that increasing the thickness of the skins strengthens the response of the sandwich structure against buckling failure, but also revealed that thicker skins are susceptible to complex failure modes. Towards data-driven constitutive modelling of the nonlinear elastic-plastic response of this bio-based structure, predictive models premised on feedforward backpropagation neural network (FFNN), cascade-forward backpropagation neural network (CFNN), and generalized regression neural network (GRNN) were developed. Performance of the models was assessed via error criteria that include the coefficient of determination (R2), root mean squared error (RMSE) and mean absolute error (MAE). Results from the models indicate that CFNN with 15 hidden neurons under the Levenberg-Marquardt backpropagation training algorithm outperformed FFNN and GRNN models, with R2=1.0, RMSE=0.0030 and MAE=0.0019.

Prediction of runoff using BPNN, FFBPNN, CFBPNN algorithm in arid watershed: A case study

Here, an endeavor has been made to predict the correspondence between rainfall and runoff and modeling are demonstrated using Feed Forward Back Propagation Neural Network (FFBPNN), Back Propagation Neural Network (BPNN), and Cascade Forward Back Propagation Neural Network (CFBPNN), for predicting runoff. Various indicators like mean square error (MSE), Root Mean Square Error (RMSE), and coefficient of determination (R2) for training and testing phase are used to appraise performance of model. BPNN performs paramount among three networks having model architecture 4-5-1 utilizing Log-sig transfer function, having R2 for training and testing is correspondingly 96.43 and 95.98. Similarly for FFBPNN, with Tan-sig function preeminent model architecture is seen to be 4-5-1 which possess MSE training and testing value 0.000483, 0.001025, RMSE training and testing value 0.02316, 0.03085 and R2 for training and testing as 0.9925, 0.9611, respectively. But for FFBPNN the value of R2 in training and testing is 0.8765 0.8976. Outcomes on the whole recommend that assessment of runoff is suitable to BPNN as contrasted to CFBPNN and FFBPNN. This consequence helps to plan, arrange and manage hydraulic structures of watershed.

A diagnostic genomic signal processing (GSP)-based system for automatic feature analysis and detection of COVID-19.

Coronavirus Disease 2019 (COVID-19) is a sudden viral contagion that appeared at the end of last year in Wuhan city, the Chinese province of Hubei, China. The fast spread of COVID-19 has led to a dangerous threat to worldwide health. Also in the last two decades, several viral epidemics have been listed like the severe acute respiratory syndrome coronavirus (SARS-CoV) in 2002/2003, the influenza H1N1 in 2009 and recently the Middle East respiratory syndrome coronavirus (MERS-CoV) which appeared in Saudi Arabia in 2012. In this research, an automated system is created to differentiate between the COVID-19, SARS-CoV and MERS-CoV epidemics by using their genomic sequences recorded in the NCBI GenBank in order to facilitate the diagnosis process and increase the accuracy of disease detection in less time. The selected database contains 76 genes for each epidemic. Then, some features are extracted like a discrete Fourier transform (DFT), discrete cosine transform (DCT) and the seven moment invariants to two different classifiers. These classifiers are the k-nearest neighbor (KNN) algorithm and the trainable cascade-forward back propagation neural network where they give satisfying results to compare. To evaluate the performance of classifiers, there are some effective parameters calculated. They are accuracy (ACC), F1 score, error rate and Matthews correlation coefficient (MCC) that are 100%, 100%, 0 and 1, respectively, for the KNN algorithm and 98.89%, 98.34%, 0.0111 and 0.9754, respectively, for the cascade-forward network.

Assessment of Sediment Load Concentration Using SVM, SVM-FFA and PSR-SVM-FFA in Arid Watershed, India: A Case Study

Improvement in area of artificial intelligence for predicting different hydrological phenomenon has shaped an enormous alteration in predictions. Knowledge on suspended sediment load (SSL) is vital in managing water resources problems and safe guard environment. Present study evaluated accurateness of five soft computing techniques, i.e. radial basis function network (RBFN), cascade forward back propagation neural network (CFBPNN), support vector machine (SVM), integration of support vector machine with firefly algorithm (SVM-FFA) and phase space reconstruction (PSR) with SVM-FFA (PSR-SVM-FFA) approaches to estimate daily SSL in Salebhata, Suktel, Lant gauge stations in western part of Odisha, India. Performance of selected models were evaluated on basis of performance criterion namely root mean square error (RMSE), Nash-Sutcliffe (NSE), Wilton index (WI) for choosing best fit model. Results acquired verified that application of various neural network methods in present field of study showed fine concurrence with observed SSL values. Comparison of estimation accuracies of different methods exemplified that PSR-SVM-FFA is very precise to estimate SSL when compared with other models. Result shows that Suktel gauge station, the best value of WI is 0.978 for PSR-SVM-FFA model, while it is 0.959, 0.923, 0.885, and 0.842 for SVM-FFA, SVM, CFBPNN, RBFN models in testing phase. Moreover, cumulative SSL data calculated by PSR-SVM-FFA method are closer to observed data as compared to other methods.