Deep Belief Network Algorithm Research Articles

Analysis of Multimedia Teaching Path of Popular Music Based on Multiple Intelligence Teaching Mode

Pop music multimedia is one of the popular digital pop music types. Based on the multiple intelligences teaching model, a multimedia multiple intelligences teaching method of popular music is proposed. This method not only analyzes the characteristics of pop music in detail, but also fully considers other important characteristics of pop music. It teaches college students multimedia, purifies their hearts, improves their personality, cultivates their innovative consciousness, and promotes their healthy growth. In this paper, the multiple intelligences teaching model is introduced into the process of pop music multimedia teaching path. In the stage of music audio segmentation, the deep belief network algorithm is used to accurately carry out music multimedia teaching. Finally, the experimental analysis results show that the integration of pop factors into music teaching, the combination of pop music and quality education, and the creation of music that is suitable for students’ personality characteristics and the needs of aesthetic development can better serve the needs of students’ quality improvement.

A precise method of color space conversion in the digital printing process based on PSO-DBN

Neural networks have been widely used in color space conversion in the digital printing process. The shallow neural network easily obtains the local optimal solution when establishing multi-dimensional nonlinear mapping. In this paper, an improved high-precision deep belief network (DBN) algorithm is proposed to achieve the color space conversion from CMYK to L*a*b*. First, the PANTONE TCX color card is used as sample data, in which the CMYK value of the color block is used as input and the L*a*b* value is used as output; then, the conversion model from CMYK to L*a*b* color space is established by using DBN. To obtain better weight and threshold, DBN is optimized by a particle swarm optimization algorithm. Experimental results show that the proposed method has the highest conversion accuracy compared with Back Propagation Neural Network, Generalized Regression Neural Network, and traditional DBN color space conversion methods. It can also adapt to the actual production demand of color management in digital printing.

Cognitive Computational Model Using Machine Learning Algorithm in Artificial Intelligence Environment

Abstract In order to explore the application of machine learning algorithm to intelligent analysis of big data in an artificial intelligence (AI) environment, make cognitive computing meet the requirements of AI and better assist humans to carry out data analysis, first, the theoretical basis of machine learning algorithm is elaborated. Then, a cognitive computational model based on the machine learning algorithm is proposed, including the essence, principle, function, training method of deep belief network (DBN) algorithm, as well as the joint use of DBN algorithm and multilayer perceptron. Finally, the proposed algorithm is simulated. The results show that under the same parameter conditions, the accuracy rate of the DBN algorithm combined with multilayer perceptron is higher than that of the DBN algorithm; when the number of units is >40, the accuracy rate of the DBN algorithm combined with multilayer perceptron is significantly higher than that of the DBN algorithm; when the number of units is 30, the best effect can be obtained, and the error rate is <0.05, but the DBN algorithm cannot achieve this effect alone; when the number of network layers is specified as four, the error rate of the DBN algorithm combined with multilayer perceptron is <0.05, forming the optimal level. In the AI environment, the performance of the cognitive computational model based on the DBN algorithm and multilayer perceptron can reach the highest level, which makes the computer become a handy intelligent auxiliary tool for human beings.

Research on Image Classification Algorithm Based on Convolutional Neural Network

Nowadays, we are in the information age. Pictures carry a lot of information and play an indispensable role. For a large number of images, it is very important to find useful image information within the effective time. Therefore, the excellent performance of the image classification algorithm has certain influence factors on the result of image classification. Image classification is to input an image, and then use a certain classification algorithm to determine the category of the image. The main process of image classification: image preprocessing, image feature extraction and classifier design. Compared with the manual feature extraction of traditional machine learning, the convolutional neural network under the deep learning model can automatically extract local features and share weights. Compared with traditional machine learning algorithms, the image classification effect is better. This paper focuses on the study of image classification algorithms based on convolutional neural networks, and at the same time compares and analyzes deep belief network algorithms, and summarizes the application characteristics of different algorithms.

A novel wavelet threshold denoising and deep belief network fault detection algorithm

For oil pipeline leakage fault detection problems, a novel negative pressure wave (NPW) leak detection method based on wavelet threshold denoising and deep belief network (Wavelet-DBN) is proposed. Firstly, the wavelet threshold denoising method is used to deal with the sample pressure signal, and the results of wavelet denoising with different wavelet basis functions and different decomposition levels are compared. The optimal parameters are selected for wavelet denoising and the characteristic information of a pipeline pressure signal is extracted. Secondly, in order to improve the accuracy of the pipeline leakage monitoring method based on NPW, the deep belief network algorithms are proposed to classify and identify the NPW sample signals. Finally, the sample data are collected from the industrial oil pipeline leakage experiment. The simulation experimental results show that the proposed method has a higher accuracy rate than other traditional machine learning methods, such as support vector machines, and reduces the false alarm rate of oil pipeline leakage monitoring.

Hybridization of DBN with SVM and its Impact on Performance in Multi-Document Summarization

Data available from web based sources has grown tremendously with growth of the internet. Users interested in information from such sources often use a search engine to obtain the data which they edit for presentation to their audience. This process can be tedious especially when it involves the generation of a summary. One way to ease the process is by automation of the summary generation process. Efforts by researchers towards automatic summarization have yielded several approaches among them machine learning. Thus, recommendations have been made on combining the algorithms with different strengths, also called hybridization, in order to enhance their performance. Therefore, this research sought to establish the impact of hybridization of Deep Belief Network (DBN) with Support Vector Machine (SVM) on precision, recall, accuracy and F-measure when used in the case of query oriented multi-document summarization. The experiments were carried out using data from National Institute of Standards and Technology (NIST), Document Understanding Conference (DUC) 2006. The data was split into training and test data and used appropriately in DBN, SVM, SVM-DBN hybrid and DBN-SVM hybrid. Results indicated that the hybridized algorithm has better precision, accuracy and F-measure as compared to DBN. Pre-classification hybridization of DBN with SVM (SVM-DBN) gives the best results. This research implies that use of DBN and SVM hybrid algorithms would enhance query oriented multi-document summarization.

Leaf area index remote sensing based on Deep Belief Network supported by simulation data

ABSTRACT Leaf area index (LAI) is a key variable in the exchange of substance and energy between the surface and the atmosphere. Remote sensing inversion is the most effective method used to obtain the LAI in a large area. However, owing to the complexity of the spatial structure of vegetation, it is difficult to obtain LAI measurements with high stability and precision. One of the main reasons for this is the lack of information mining ability provided by remote sensing images. To fully utilize the information provided by these images, deep learning technology, with strong self-learning ability and information mining ability, was proposed in this study to retrieve the LAI. The accuracy and stability of deep learning technology largely depends on the quality, quantity, and representativeness of the samples. Given the present difficulties in producing enough high-quality samples from land surface measurements, this paper proposes the use of a radiative transfer model to simulate samples to realize a remote sensing inversion of the LAI. The PROSAIL model is used to simulate the training samples for LAI inversion. A Deep Belief Network (DBN) was used for LAI inversion from MODIS (Moderate-Resolution Imaging Spectroradiometer) data with seven spectral bands, and the estimated LAI was compared with the current MODIS LAI product (MOD15A2H), on the basis of validation by ground-measured LAI. The inversion results (Root Mean Square Error RMSE and Pearson’s correlation coefficient r) obtained by the DBN algorithm (RMSE = 0.8988, r = 0.7188) of this study and GLASS (Global LAnd Surface Satellite) algorithm (RMSE = 0.7111, r = 0.7995) showed a similar performance, and they are superior to the MODIS LAI product (RMSE = 1.0595, r = 0.6613).

Fault diagnosis of railway freight car wheelset based on deep belief network and cuckoo search algorithm

As the key running part of the railway freight transportation system, the wheel not only bears the load of the vehicle, but also ensures the running and steering of the car body on the rails. The frequent high-speed friction with the rail and brake is the main reason for early failure of wheelset tread. Therefore, real-time status monitoring and early fault diagnosis of wheel treads have become key technical issues that must be solved in the reform of the railway freight maintenance system. In this paper, an adaptive hybrid Simulated Annealing Cuckoo Search algorithm (SA-ACS) is proposed and applied to the Deep Belief Network (DBN). The SA-ACS-DBN algorithm is used to improve the training speed and convergence accuracy of the diagnosis model. Finally, it is found through the comparison experiment of wheel tread fault data that the data results prove the feasibility of the SA-ACS-DBN model in the application of wheelset fault diagnosis.

Recognition of Flexion and Extension Imagery Involving the Right and Left Arms Based on Deep Belief Network and Functional Near-Infrared Spectroscopy.

Brain-computer interaction based on motor imagery (MI) is an important brain-computer interface (BCI). Most methods for MI classification are based on electroencephalogram (EEG), and few studies have investigated signal processing based on MI-Functional Near-Infrared Spectroscopy (fNIRS). In addition, there is a need to improve the classification accuracy for MI fNIRS methods. In this study, a deep belief network (DBN) based on a restricted Boltzmann machine (RBM) was used to classify fNIRS signals of flexion and extension imagery involving the left and right arms. fNIRS signals from 16 channels covering the motor cortex area were recorded for each of 10 subjects executing or imagining flexion and extension involving the left and right arms. Oxygenated hemoglobin (HbO) concentration was used as a feature to train two RBMs that were subsequently stacked with an additional softmax regression output layer to construct DBN. We also explored the DBN model classification accuracy for the test dataset from one subject using training dataset from other subjects. The average DBN classification accuracy for flexion and extension movement and imagery involving the left and right arms was 84.35 ± 3.86% and 78.19 ± 3.73%, respectively. For a given DBN model, better classification results are obtained for test datasets for a given subject when the model is trained using dataset from the same subject than when the model is trained using datasets from other subjects. The results show that the DBN algorithm can effectively identify flexion and extension imagery involving the right and left arms using fNIRS. This study is expected to serve as a reference for constructing online MI-BCI systems based on DBN and fNIRS.

Diagnosis analysis of rectal function through using ensemble empirical mode decomposition-deep belief networks algorithm.

The rectal motility function can reflect a person's rectal health status. To diagnose the rectal motility function after artificial anal sphincter implantation, this paper proposes a rectal function diagnosis model based on ensemble empirical mode decomposition-deep belief networks (EEMD-DBNs). Because of the rectal pressure signals that are unstable and subjected to noise interferences, an EEMD framework based on EMD, which can reduce the effect of signal modal mixing, is proposed. EMD and EEMD were used to decompose the analog signal, respectively, and it was found that EEMD can significantly reduce the effect of mode aliasing. During the rectal pressure signal decomposition experiment, by analyzing the intrinsic mode functions generated by the signals from normal people and diseased patients, the rectal signals at these two different conditions can be well distinguished. Additionally, the DBN was introduced to perform deep learning to extract the multi-dimensional features of rectal signals and then output the classification results via using the top-level classifier, which can overcome the difficulties in extracting the rectal signal features. The results showed that, following the principle of balancing the diagnosis accuracy and model running speed, the best diagnosis performance was achieved when three restricted Boltzmann machines and five layers of DBN model were set, with the diagnosis rate of 85%. The diagnostic model used in this study can distinguish the signals between normal and abnormal rectal functions with accurate performance, thus providing the technical support for the recovery of the rectal motility function of artificial anal sphincter implanters.

Role Comparison between Deep Belief Neural Network and NeuroEvolution of Augmenting Topologies to Detect Diabetes

Diabetes in Indonesia has been perceived as a grave health problem and has been a concern since the early 1980’s [2]. The prevalence of diabetes in adults in Indonesia, as stated by IDF, was 6.2% with the total case amounting to 10.681.400. Moreover, Indonesia is also in the top ten global countries with the highest diabetes case in 2013. This research will investigate the role of Deep Belief Network (DBN) and NeuroEvolution of Augmenting Topology (NEAT) in solving regression problems in detecting diabetes. DBN works by processing the data in unsupervised network architectures. The algorithm puts Restricted Boltzmann Machines (RBM) into a stacked process. The output of the first RBM will be the input for the next RBM. On the other hand, the NEAT algorithm works by investigating the neural network architecture and evaluating the architecture using a multi-layer perceptron algorithm. Collaboration with a Genetic Algorithm in NEAT is the key process in architecture development. The research results showed that DBN could be utilized as the initial weight for Backpropagation Neural Network at 22.61% on average. On the other hand, the NEAT algorithm could be used by collaborating with a multi-layer perceptron to solve this regression problem by providing 74.5% confidence. This work also reveals potential works in the future by combining the Backpropagation algorithm with NEAT as an evaluation function and by combining it with DBN algorithms to process the produced initial weight.

Tunnel boring machines (TBM) performance prediction: A case study using big data and deep learning

This work explores the potential for predicting TBM performance using deep learning. It focuses on a 17.5-km-long tunnel excavated for the Yingsong Water Diversion Project in Northeastern China with its 728 days’ continuous monitoring of mechanical data. The prediction uses the deep belief network (DBN) proposed by Hinton et al. (2006),on the penetration rate, cutter rotation speed, torque, and thrust force. Field Penetration Index (FPI) is introduced to quantify TBM performance in the field. The DBN algorithm trains on nth number of preceding elements and predicts the performance of the n + 1th element. Prior to the implementation of the DBN, a pilot test was performed to find the optimal values for the network structural parameters (number of input nodes, number of hidden layers, number of nodes in the hidden layers, and learning rate). Predictions on FPIs in all the three rock types were then proceeded with good agreement with the field measured data. The mean relative errors for the predicted measured FPIs are generally less than 0.15 and the correlation coefficients (R) can be higher than 0.78. The predicted and measured FPI values along the length of the tunnel graphically follow the same trends. These results confirm the usefulness of big data and the deep learning in predicting TBM performance.

The use of animal sensor data for predicting sheep metabolisable energy intake using machine learning

The use of sensors for monitoring livestock has opened up new possibilities for the management of livestock in extensive grazing systems. The work presented in this paper aimed to develop a model for predicting the metabolisable energy intake (MEI) of sheep by using temperature, pitch angle, roll angle, distance, speed, and grazing time data obtained directly from wearable sensors on the sheep. A Deep Belief Network (DBN) algorithm was used to predict MEI, which to our knowledge, has not been attempted previously. The results demonstrated that the DBN method could predict the MEI for sheep using sensor data alone. The mean square error (MSE) values of 4.46 and 20.65 have been achieved using the DBN model for training and testing datasets, respectively. We also evaluated the influential sensor data variables, i.e., distance and pitch angle, for predicting the MEI. Our study demonstrates that the application of machine learning techniques directly to on-animal sensor data presents a substantial opportunity to interpret biological interactions in grazing systems directly from sensor data. We expect that further development and refinement of this technology will catalyse a step-change in extensive livestock management, as wearable sensors become widely used by livestock producers.

Deep Learning Based Semi-Supervised Control for Vertical Security of Maglev Vehicle With Guaranteed Bounded Airgap

The vertical security problem of maglev train is challenging for nonlinearity, external disturbances, unmeasurable airgap velocity and constrained output. To solve this problem, a semi-supervised controller based on deep belief network (DBN) algorithm is proposed in the presence of unknown external disturbances. Firstly, the extended state observer (ESO) is designed to ensure fast convergence of observation errors with high enough estimation precision. An output-constrained controller is designed by backstepping method, and the estimated value of ESO is introduced to ensure that the output airgap is constrained within a bounded range. Then, the stability of this method is proved based on the symmetric Barrier Lyapunov function. Subsequently, a semi-supervised controller is presented based on DBN algorithm and the output-constrained controller. The numerical simulation results show that this method can effectively deal with unmeasurable airgap velocity and generalized external disturbances, and guarantee the vertical security with output airgap within a bounded range. Finally, experiments are implemented on a full-scale maglev vehicle and the experimental results demonstrate that the developed deep learning controller can ensure the vertical security.

Research on Information Visualization Graphic Design Teaching Based on DBN Algorithm.

With the advent of the era of big data, how to quickly obtain effective information and efficiently disseminate information technology has become the most popular topic. Studies have shown that the ability of the human brain to process data and information is unmatched by machines, and the processing of graphics is tens of thousands of times faster than that of words. Based on the deep belief network (DBN) algorithm, this paper studies the technology of information visualization graphic design teaching application. Firstly, the structure of the deep belief network is analysed to explore its technical application in graphic information reconstruction. It is concluded that the DBN algorithm can be used to deal with the problems of classification, regression, dimension calculation, feature point acquisition, accuracy calculation, and so on in machine learning training. Then, the deformation technology of graphic local design is studied based on the DBN algorithm to construct the visual teaching platform and analyse the technical research results of this algorithm in information graphic design. The results show that the DBN algorithm can quickly solve the problem of processing complex features in graphics, change the local deformation design of the original graphics to form new feature point data and add it to the teaching platform, and improve the ability of model fast learning and training, optimizing the operation efficiency of the teaching platform.

Research on the hull form optimization using the surrogate models

The ship hull form optimization using the Computational Fluid Dynamics (CFD) method is increasingly employed in the early design of a ship, as an optimal ship hull form can obtain good hydrodynamics. However, it is time-consuming due to its many CFD simulations for the optimization. This paper presents a ship hull form optimization loop using the surrogate model, deep belief network (DBN), to reduce the wave-making resistance of the Wigley ship. The prediction performance of the wave-making resistance of the Wigley ship using the DBN method is discussed and compared with the traditional surrogate models found in this study. The results show that the resistance obtained using the deep belief network algorithm is superior to that obtained using the typical surrogate models. Then, a ship hull form optimization framework is built by integrating the Free From Deformation, non-linear programming by quadratic Lagrangian and deep belief network algorithms. The optimization results show that the deep belief network-based ship hull form optimization loop can be used to optimize the Wigley ship. The study presented in this paper could provide a deep learning algorithm for the ship design optimization.

Deep Belief Network with FOA based Cooperative Spectrum Sensing in Cognitive Radio Network

The artificial neural network (ANN) is proposed to the cooperative spectrum sensing (CSS) at CR network. ANN has the drawback that the training of ANN with many hidden layers on large amount of data can affect the performance of the network and optimisation of ANN parameters is a challenging task. To overcome the above drawbacks, the deep belief network (DBN) and fruit fly optimisation algorithm (FOA) are employed. The DBN has four parameters on learning step: learning rate, weight decay, penalty parameter, number of hidden units. These parameters should be properly selected for the proper functioning of DBN. Tuning of these parameters is taken into an optimisation issue and it is addressed by FOA. The proposed method has three steps: training, validation and testing. The metrics used for the performance evaluation are accuracy, false alarm rate and loss detection.

Flash flood susceptibility mapping using a novel deep learning model based on deep belief network, back propagation and genetic algorithm

Flash floods are responsible for loss of life and considerable property damage in many countries. Flood susceptibility maps contribute to flood risk reduction in areas that are prone to this hazard if appropriately used by land-use planners and emergency managers. The main objective of this study is to prepare an accurate flood susceptibility map for the Haraz watershed in Iran using a novel modeling approach (DBPGA) based on Deep Belief Network (DBN) with Back Propagation (BP) algorithm optimized by the Genetic Algorithm (GA). For this task, a database comprising ten conditioning factors and 194 flood locations was created using the One-R Attribute Evaluation (ORAE) technique. Various well-known machine learning and optimization algorithms were used as benchmarks to compare the prediction accuracy of the proposed model. Statistical metrics include sensitivity, specificity accuracy, root mean square error (RMSE), and area under the receiver operatic characteristic curve (AUC) were used to assess the validity of the proposed model. The result shows that the proposed model has the highest goodness-of-fit (AUC = 0.989) and prediction accuracy (AUC = 0.985), and based on the validation dataset it outperforms benchmark models including LR (0.885), LMT (0.934), BLR (0.936), ADT (0.976), NBT (0.974), REPTree (0.811), ANFIS-BAT (0.944), ANFIS-CA (0.921), ANFIS-IWO (0.939), ANFIS-ICA (0.947), and ANFIS-FA (0.917). We conclude that the DBPGA model is an excellent alternative tool for predicting flash flood susceptibility for other regions prone to flash floods.

Semisupervised Graph Convolution Deep Belief Network for Fault Diagnosis of Electormechanical System With Limited Labeled Data

The labeled monitoring data collected from the electromechanical system is limited in the real industries; traditional intelligent fault diagnosis methods cannot achieve satisfactory accurate diagnosis results. To deal with this problem, an intelligent fault diagnosis method for electromechanical system based on a new semisupervised graph convolution deep belief network algorithm is proposed in this article. Specifically, the labeled and unlabeled samples are first employed to design a new adaptive local graph learning method for constructing the graph neighbor relationship. Meanwhile, the labeled samples are applied to describe the discriminative structure information of data via the latest circle loss. Finally, the local and discriminative objective functions are reconstructed under the semisupervised learning framework. The experimental results from the motor-bearing system demonstrate that the method can achieve 98.66 % accuracy with only 10 % of training labeled data, which indicates that it is a promising semisupervised intelligent fault diagnosis method.

Smartphone Motion Sensor-Based Complex Human Activity Identification Using Deep Stacked Autoencoder Algorithm for Enhanced Smart Healthcare System.

Human motion analysis using a smartphone-embedded accelerometer sensor provided important context for the identification of static, dynamic, and complex sequence of activities. Research in smartphone-based motion analysis are implemented for tasks, such as health status monitoring, fall detection and prevention, energy expenditure estimation, and emotion detection. However, current methods, in this regard, assume that the device is tightly attached to a pre-determined position and orientation, which might cause performance degradation in accelerometer data due to changing orientation. Therefore, it is challenging to accurately and automatically identify activity details as a result of the complexity and orientation inconsistencies of the smartphone. Furthermore, the current activity identification methods utilize conventional machine learning algorithms that are application dependent. Moreover, it is difficult to model the hierarchical and temporal dynamic nature of the current, complex, activity identification process. This paper aims to propose a deep stacked autoencoder algorithm, and orientation invariant features, for complex human activity identification. The proposed approach is made up of various stages. First, we computed the magnitude norm vector and rotation feature (pitch and roll angles) to augment the three-axis dimensions (3-D) of the accelerometer sensor. Second, we propose a deep stacked autoencoder based deep learning algorithm to automatically extract compact feature representation from the motion sensor data. The results show that the proposed integration of the deep learning algorithm, and orientation invariant features, can accurately recognize complex activity details using only smartphone accelerometer data. The proposed deep stacked autoencoder method achieved 97.13% identification accuracy compared to the conventional machine learning methods and the deep belief network algorithm. The results suggest the impact of the proposed method to improve a smartphone-based complex human activity identification framework.