Deep Belief Network Research Articles

The Novel Meta-Heuristic Optimization-Based Deep Learning Framework for Predicting Diagnoses with COVID-19 in Patients Undergoing Emergency Treatment

The global medical system has faced enormous challenges as a result of the COVID-19 epidemic. Since emergency facilities are frequently the first places individuals with COVID-19 indications go, they are often in the center of the matter. Finding patients with COVID-19 and treating them appropriately while applying precautions to stop the virus from spreading among other individuals and medical personnel is one of the primary challenges EDs face. This work develops a Covid-19 forecasting system using deep learning via four essential steps. The incoming initial information is initially placed via the pre-processing phase to improve the information accuracy and efficacy evaluation of the suggested model. Data cleansing and normalization are done during the pre-processing phase. The best characteristics are chosen using meta-heuristic-based Belief Net Particle Swarm optimization (MH-Belief Net + PSO). Next, the covid-19 forecasting step is replicated using the newly improved Deep Learning (DL) approach, the optimizing deep belief network (DBN). The parameter modification enhances the system's capacity to forecast disease. An improved DBN's output shows if COVID-19 is present or nonexistent. Because of this, the effectiveness assessment significance of the suggested approaches was greater compared to each of existing approaches, including SVM, RF, CNN, and NB.

An adaptive deep belief network for fault diagnosis of complex analog circuits

An adaptive deep belief network for fault diagnosis of complex analog circuits

A prognostic model for multi-stage degraded equipment under zero life label combining CDBN and Bayesian bidirectional GRU

Predicting remaining useful life (RUL) of multi-stage degraded equipment under the big data era is generally implemented by the following two steps: degradation stage division and RUL prediction by stage. However, three aspects of problems may be encountered, such as zero life label resulting from censored data, degradation stage number dependent on empirical knowledge, and prediction results ignoring the uncertainty of stage results, which leads to poor prediction effect inevitably. Towards this end, this paper proposes a prognostic model for multi-stage degraded equipment under zero life label. Firstly, a degradation stage division model for multi-stage degraded equipment is constructed and trained based on continuous deep belief network (CDBN). Next, we employ the sliding time window to generate the samples with the degradation label, and utilize Bayesian bidirectional gated recurrent unit (GRU) to describe the performance evolution model for each stage. For the concerned equipment in service, the current multiple domain features are input into the stage division model to compute the probability of various stages. Furthermore, the degradation feature can be extrapolated by the prediction model of each stage respectively. As a result, we can obtain the predicted degradation distribution according to the law of total probability. Similarly, the degradation level at any time can be calculated by treating the last prediction as the monitoring information in the iterative manner, and the RUL distribution can be determined combining with the failure threshold. Finally, we resort to a case study to verify the effectiveness of our prognostic model.

Clustering Molecules at a Large Scale: Integrating Spectral Geometry with Deep Learning.

This study conducts an in-depth analysis of clustering small molecules using spectral geometry and deep learning techniques. We applied a spectral geometric approach to convert molecular structures into triangulated meshes and used the Laplace-Beltrami operator to derive significant geometric features. By examining the eigenvectors of these operators, we captured the intrinsic geometric properties of the molecules, aiding their classification and clustering. The research utilized four deep learning methods: Deep Belief Network, Convolutional Autoencoder, Variational Autoencoder, and Adversarial Autoencoder, each paired with k-means clustering at different cluster sizes. Clustering quality was evaluated using the Calinski-Harabasz and Davies-Bouldin indices, Silhouette Score, and standard deviation. Nonparametric tests were used to assess the impact of topological descriptors on clustering outcomes. Our results show that the DBN + k-means combination is the most effective, particularly at lower cluster counts, demonstrating significant sensitivity to structural variations. This study highlights the potential of integrating spectral geometry with deep learning for precise and efficient molecular clustering.

Fusion of deep belief network and SVM regression for intelligence of urban traffic control system

Fusion of deep belief network and SVM regression for intelligence of urban traffic control system

Ensemble learning of decomposition-based machine learning models for multistep-ahead daily streamflow forecasting in northwest China

ABSTRACT Accurate daily streamflow forecasts remain challenging in arid regions. A Bayesian model averaging (BMA) ensemble learning strategy was proposed to forecast 1-, 2-, and 3-day-ahead streamflow in Dunhuang Oasis, northwest China. The efficiency of BMA was compared with four decomposition-based machine learning and deep learning models. Satisfactory forecasts were achieved with all proposed models at all lead times; however, based on Nash-Sutcliffe efficiency values of 0.976, 0.967, and 0.957, BMA achieved the greatest accuracy for 1-, 2-, and 3-day-ahead streamflow forecasts, respectively. Uncertainty analysis confirmed the reliability of BMA in yielding consistently accurate streamflow forecasts. Thus, BMA could provide an efficient alternative approach to multistep-ahead daily streamflow forecasting. The incorporation of data decomposition techniques (e.g. variational mode decomposition) and deep learning algorithms (e.g. deep belief network) into BMA may provide worthy technical references for supervised learning of streamflow systems in data-scarce regions.

Exploring role of prefrontal cortex region of brain in children having ADHD with machine learning: Implications and insights

Objective Attention deficit hyperactivity disorder (ADHD), is a general neurodevelopmental syndrome. This affects both adults and children, causing issues like hyperactivity, inattention, and impulsivity. Diagnosis, typically reliant on patient narratives and questionnaires, can sometimes be inaccurate, leading to distress. We propose utilizing empirical mode decomposition (EMD) for feature extraction and a machine learning (ML) algorithm to categorize ADHD and control. Method Publicly available Kaggle dataset is used for research. The EMD technique decomposes an electroencephalogram (EEG) waveform to 12 intrinsic mode functions (IMFs). Thirty-one statistical parameters are generated over the first 6 IMFs to create an input feature vector for the deep belief network (DBN) classifier. Principal component analysis (PCA) is utilized to reduce dimension. Findings Experimental results are compared on prefrontal cortex channels Fp1 and Fp2. After an in-depth evaluation of all metrics, it is observed that, in patients with ADHD, the prefrontal cortex regulates attention, behavior, and emotion. Our findings align with established neuroscience. The critical functions of the brain, such as organization, planning, attention, and decision making, are performed by the frontal lobe. Novelty Our work provides a novel approach to understanding the disorder’s underlying neurobiological mechanisms. It has the potential to deepen our understanding of the condition, improve diagnostic accuracy, personalize treatment methods, and, ultimately, improve outcomes for those affected.

Degradation modeling and remaining life prediction for a multi-component system under triple uncertainties

As equipment systems are gradually moving toward large-scale and complex that comprise multi-components, the correlation between components significantly impacts the system’s degradation. Meanwhile, the system degradation process is subject to aleatory uncertainty due to inherent fluctuations, epistemic uncertainty since insufficient information, and measurement uncertainty affected by measurement noise. In this paper, a system-level RUL prediction method based on deep belief network, self-organizing map neural network and uncertain random process is proposed by combining the advantages of deep learning in extracting degraded features with the advantages of statistical-based models in quantifying uncertainty. Firstly, the monitoring variables that characterize the equipment degradation trend are selected based on multi-sensor monitoring information of a single component. Deep belief network is utilized to extract component-level deep hidden degradation features. Subsequently, a system-level health index construction method combining mutual information and self-organizing map neural network is proposed according to correlation analysis between components. Finally, a health index evolution model is constructed to predict system RUL by integrating the advantages of the multi-phase degradation model based on uncertain random processes quantifying multiple uncertainties. The superiority of the proposed method is verified through the C-MAPSS dataset provided by NASA.

Defence against adversarial attacks on IoT detection systems using deep belief network

An Adversarial attack is a technique used to deceive machine learning models to make incorrect predictions by providing slightly modified inputs from the original. Intrusion detection system (IDS) is a crucial tool in computer network security for the detection of adversarial attacks. Deep learning is a trending method in both research and industry, and this study proposes the use of a deep belief network (DBN). DBN can recognize data with small differences, but is also vulnerable to adversarial attacks. Therefore, this research suggests an internet of things-intrusion detection system (IoT-IDS) architecture using a DBN that can counter adversarial attacks. The chosen adversarial attack for this study is the fast gradient sign method (FGSM) used to evaluate the IoT IDS using the DBN model. Testing was conducted in two scenarios: first, the model was trained without adversarial attacks; second, the model was trained with adversarial attacks. The test results indicate that the DBN model struggles to detect FGSM attacks, achieving an accuracy of only 46% when it is not trained with adversarial attacks. However, after training with the FGSM dataset, the DBN model successfully detected adversarial attacks with an accuracy of 97%.

Malignant thyroid lump multi classification by TIRADS using DBA with transfer learning

Thyroid diseases have developed into significant illnesses in recent decades. These diseases affect the thyroid glands and are caused by elevated thyroid hormone levels or infections in the thyroid organs. It is challenging to resolve thyroid diagnosis using conventional parametric and nonparametric statistical techniques since it can be viewed as a classification problem. However, there are certain barriers in the manner of obtaining both efficacy and accuracy in thyroid nodule diagnosis. Deep learning (DL) and machine learning (ML) models have emerged as useful instruments for the diagnosis of sickness in the modern era. For the purpose of diagnosing and classifying thyroid diseases, this research introduces a novel deep belief network (DBF) with transfer learning, known as DBNTL. In this study, the pre-processed image was first pre-processed using a conventional multiresolution bilateral technique, and then it was subjected to a novel segmentation technique called fusion pooling integrated U-net segmentation. The DBN with transfer learning model is used to classify and grade malignant thyroid nodules in compliance with thyroid imaging-reporting-and-data-system (TIRADS) guidelines. In this model, the model's weights are obtained by transfer learning. A major metric for evaluating the efficacy of biological image processing applications, good sensitivity and specificity (97.28 and 97.22, respectively) were obtained for the recommended modes.

Network Intrusion Detection Based on Deep Belief Network Broad Equalization Learning System

Network intrusion detection systems are an important defense technology to guarantee information security and protect a network from attacks. In recent years, the broad learning system has attracted much attention and has been introduced into intrusion detection systems with some success. However, since the traditional broad learning system is a simple linear structure, when dealing with imbalanced datasets, it often ignores the feature learning of minority class samples, leading to a poorer recognition rate of minority class samples. Secondly, the high dimensionality and redundant features in intrusion detection datasets also seriously affect the training time and detection performance of the traditional broad learning system. To address the above problems, we propose a deep belief network broad equalization learning system. The model fully learns the large-scale high-dimensional dataset via a deep belief network and represents it as an optimal low-dimensional dataset, and then introduces the equalization loss v2 reweighing idea into the broad learning system and learns to classify the low-dimensional dataset via a broad equalization learning system. The model was experimentally tested using the CICIDS2017 dataset and fully validated using the CICIDS2018 dataset. Compared with other algorithms in the same field, the model shortens the training time and has a high detection rate and a low false alarm rate.

Ammonia emission concentration measurement using laser absorption spectroscopy chamber system based on deep belief networks

Ammonia emission concentration measurement using laser absorption spectroscopy chamber system based on deep belief networks

Retraction Note: Convolution neural network and deep-belief network (DBN) based automatic detection and diagnosis of Glaucoma

Retraction Note: Convolution neural network and deep-belief network (DBN) based automatic detection and diagnosis of Glaucoma

Automated English Teaching System Through Deep Belief Network for Human-Computer Interaction Experience

This paper presented the integration of Human-Computer Interaction (HCI) with the Automated Teaching Belief Network (ATBN) to enhance automated English teaching experiences. The proposed ATBN model implemented the Deep Belief Network for the estimation of the factors related to the HCI to promote the experience of the users. The ATBN model uses the deep learning model for the classification in English Teaching. Through the capabilities of deep learning and HCI principles, the ATBN system offers personalized and adaptive learning experiences tailored to individual student needs. The proposed ATBN model estimates the features in English teaching to improve the performance of the Students through HCI model. Simulation analysis expressed that proposed ATBN model improves the pre-test and post-test score by +15 for the English Teaching. The classification values are achieved with accuracy value of 94.8% with minimal loss of 0.12. The assessment of student performance through pre-test and post-test score is improved by 15 for the beginner, intermediate and advanced level. The findings expressed that proposed ATBN model achieves the higher teaching test performance for the HCI language level through the belief network those significantly improves the user experience.

Tracing the origin of isatidis radix based on multivariate data fusion combined with DBN classification algorithm

In this study, multidimensional characterization data such as chromaticity value, texture and compositional content of Isatidis Radix from different regions (Anhui; Hubei; Shaanxi; Xinjiang) were collected. By multivariate statistical analysis, 44 characterization factors (VIP >1, P < 0.05) were selected to distinguish the origin of Isatidis Radix. In addition, a unique artificial intelligence algorithm was created and optimized by merging 44 characterization factors with the deep belief network (DBN) classification algorithm. Compared with the traditional discriminant analysis method, the accuracy of this new method was significantly improved, and the discrimination rate of Isatidis Radix origin reached 100 %, and the traceability accuracy of Isatidis Radix also reached 100 %. This study supports the development of intelligent algorithms based on data fusion to track the origin of more agricultural products.

Complete Ensemble Empirical Mode Decomposition with Adaptive Noise to Extract Deep Information of Bearing Fault in Steam Turbines via Deep Belief Network

Extracting to enhance the accuracy of diagnosing bearing faults in steam turbines, a novel approach focused on extracting key fault features from vibration signals is introduced. Recognizing the complex, non-linear, and non-stationary nature of bearing vibration signals, our strategy involves a sensitivity analysis utilizing a multivariate diagnostic algorithm. The process begins with collecting vibration data from defective bearings via the TMI system. This data is then subjected to Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), enabling the integration of adaptive noise for the extraction of in-depth information. Following this, an analysis in both time and frequency domains — post Fast Fourier Transform (FFT) — is conducted on the decomposed signals, forming the basis of a diagnostic features database. To streamline data analysis and boost the model’s computational efficiency, a combination of eXtreme Gradient Boosting (XGBoost) and Mutual Information Criterion (MIC) is applied for dimensionality reduction. Furthermore, a deep belief network (DBN) is implemented to develop a precise fault diagnosis model for the bearings in rotating machinery. By incorporating sensitivity analysis, a diagnostic matrix is crafted, facilitating highly accurate fault identification. The superiority of this diagnostic algorithm is corroborated by testing with real on-site data and a benchmark database, demonstrating its enhanced diagnostic capabilities relative to other feature selection techniques.

Optimized Deep Belief Network for Efficient Fault Detection in Induction Motor

Numerous industrial applications depend heavily on induction motors and their malfunction causes considerable financial losses. Induction motors in industrial processes have recently expanded dramatically in size, and complexity of defect identification and diagnostics for such systems has increased as well. As a result, research has concentrated on developing novel methods for the quick and accurate identification of induction motor problems.In response to these needs, this paper provides an optimised algorithm for analysing the performance of an induction motor. To analyse the operation of induction motors, an enhanced methodology on Deep Belief Networks (DBN) is introduced for recovering properties from the sensor identified vibration signals. Restricted Boltzmann Machine (RBM) is stacked utilizing multiple units of DBN model, which is then trained adopting Ant colony algorithm.An innovative method of feature extraction for autonomous fault analysis in manufacturing is provided by experimental investigations utilising vibration signals and overall accuracy of 99.8% is obtained, which therefore confirms the efficiency of DBN architecture for features extraction.

Aquila Optimized Fuzzy Deep Belief Network for Secure Data Transmission in WSN

A wireless sensor network (WSN) is made up of several independent sensor nodes that are able to interpret, analyze, and work with data. It is generally recognized that security and limited energy are the two challenging tasks with WSNs. To address these challenges, a novel Aquila-optimized fuzzy deep belief network (AO-FDBN) model has been proposed in this paper. The suggested AO-FDBN framework consists of three stages. Initially, the Aggregator has been selected by using the Aquila optimization algorithm. Secondly, the data from the aggregator are encrypted by using the blowfish algorithm. Finally, the optimal route has been selected by using the fuzzy-deep belief network (DBN). Packet delivery ratio (PDR), transport delay, energy usage, and network lifetime are evaluated between the suggested framework and current methods. Experimental results specify that the suggested AO-FDBN approach achieves higher performance of 22.617%, 14.22%, and 15.64% than TEEFCA, HESC, and SEPC methods. This system is more effective and secure for real-time applications.

A deep belief network-based energy consumption prediction model for water source heat pump system

To achieve optimal energy efficiency in buildings, accurately forecasting the energy consumption of air conditioning systems is crucial. This study develops an energy consumption prediction model based on a deep belief network, which is constructed according to the principles of a restricted Boltzmann machine. Actual experimental data from a water source heat pump system are collected, and feature variables are selected. The study discusses the impact of model parameters and training set sizes on the performance of energy consumption prediction model. Additionally, the trend in model prediction performance is analyzed through parameter adjustments. The results show that the coefficient of determination (R2) for the optimized model has increased to 0.585. The mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE) have been reduced to 6.311, 2.512, and 1.625, respectively. The deep belief network energy consumption prediction model outperforms other common machine learning models for water source heat pump systems.

Application of a Dirichlet Distribution-Based Ensemble Surrogate Model in Aerodynamic Optimization

Surrogate models have been widely applied in the aerodynamic optimization of aircrafts, whereas the traditional individual surrogate models have the defects of low robustness and applicability. In this study, a novel ensemble surrogate model is proposed and applied in the multi-objective optimization of the airfoil. The backpropagation neural network, deep belief network, and kriging surrogate models are selected as the member surrogate models, and the Dirichlet distribution strategy is introduced to adaptively generate the weights of the member surrogate models in constructing the ensemble surrogate model. An improved multi-objective particle swarm optimization (MOPSO) framework is established by employing the α-stable distribution function to enhance the global convergence rate of the algorithm. Based on the improved MOPSO framework in which the ensemble surrogate model is embedded, the multi-objective optimization of the airfoil is conducted. The results indicate that the proposed ensemble-surrogate-model-based optimization obtains better aerodynamic performance of the airfoil under multiple operating conditions, compared to the individual-surrogate-model-based optimization.