Search

Copyright of Science & Technology of Cereals, Oils & Foods is the property of Science & Technology of Cereals, Oils & Foods Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

There is a large amount of unstructured data in the current corn cross-border supply chain system and it has the characteristics of multi-source heterogeneous. Traditional risk early warning methods have defects such as over-reliance on manual decision-making and low accuracy of early warning. In order to solve the above problems, this paper proposed a system risk early warning method of corn cross-border supply chain based on deep belief network and multi-class fuzzy support vector machine. Firstly, based on the principle of embedding coding and normalization, a large number of unstructured data in the corn cross-border supply chain system were preprocessed and converted into structured data for subsequent calculation. Then, based on the deep belief network, the high-latitude features of the data were extracted, and the change trend and correlation of risk indicators in the corn cross-border supply chain system were adaptively mined. Finally, the extracted high-dimensional features were input into the multi-class fuzzy support vector machine model for training to realize the risk classification early warning of corn cross-border supply chain. The accuracy of the algorithm proposed in this paper can reach 94.88% under the condition of similar running time. It is 52.17% higher than that of the worst algorithm, and the comprehensive performance was superior to other algorithms which can provide theoretical support for the application of system risk regulation of corn cross-border supply chain.

An informed search strategy based on random statistical analysis was developed for synchronous recognition of groundwater pollution source information and aquifer parameters. An informed search iterative course (ISIC) was accordingly designed, and each iteration included the determination of attempt point and state transition. In this paper, two improvement techniques were first adopted for choosing attempt points and judging state transition in ISIC to improve search efficiency and precision. The first improvement was that the variable radius free search method was applied to choosing the attempt point, and the size of the search radius was constantly adjusted in ISIC, taking the search ergodicity and efficiency into account. The second improvement technique was a Tsallis formula used for state transition judgment, and the controlled factor in the Tsallis formula was regulated continuously so that the search could consider ergodicity and efficiency simultaneously. Furthermore, frequent calls to the groundwater pollution numerical simulator to calculate the likelihood have inflicted a huge computational burden during ISIC. An effective way is to construct a substitute for emulating the simulator with a low calculating load. However, the mapping relation between the import and export of the numerical simulator was complex and had many variables. The precision of the substitute based on shallow learning is low sometimes. Therefore, we adopted the deep learning method and built an improved deep confidence network (DCN) substitute to emulate the highly nonlinear simulator. Finally, the synchronous recognition results for groundwater pollution source information and aquifer parameters were gained when ISIC ceased. The above-mentioned methods were verified in a case involving groundwater pollution. The consequence indicated that the ISIC with an improved DCN substitute can synchronously recognize groundwater pollution source information and aquifer parameters with a high degree of precision and efficiency. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Machine learning plays an increasingly important role in the field of disease risk prediction due to its optimization advantages. This paper proposes deep confidence network optimization based on the early warning model constructed by the neural network and chooses the restricted Boltzmann machine and backpropagation algorithm as the theoretical basis of deep confidence network construction. The deep confidence network is established through the construction and stacking of RBM, and backpropagation is used to fine-tune the network parameters to generate the model. Combined with the incidence rate data of sports injuries of physical education majors and the injury classification of the deep confidence network algorithm, the test data application results verify that the algorithm has a good effect of early warning in case of sports injuries. The survey data showed that the incidence of sports injuries was 228%, and the main risk indicators causing sports injuries were not drinking alcohol (95.31%) and incorrect sports knowledge (92.09%). The model correctly predicted 94.15% (95% CI: 0.9204, 0.9608) with sensitivity and specificity: 0.954 and 0.923, respectively.

As data science and computer science continue to advance, the application of deep learning methods to address technological innovation and optimize interior design has emerged as a significant area of research. In this paper, we propose a VR matching algorithm that utilizes deep convolutional networks to match the appropriate visual elements of the room layout, thereby realizing a panoramic scheme for interior design. A method of automatic extraction of features and labels of room layout based on a convolutional neural network is designed, trained, and tested with the help of a convolutional deep confidence network, and the labeled room layout feature map is classified using the Softmax classifier. A fast retrieval algorithm is being investigated with input in the form of images and combining text and content. Firstly, we input the room layout feature image, extract the feature vector using a convolutional neural network model, and then calculate the similarity between the feature vector in the visual element decorative design drawing and the feature vector of the input room layout image. The maximum similarity is selected as the visual element design scheme of the feature, and then the picture of the panoramic scheme is realized in order to let the designer get the customer’s feedback on the design effect in a more timely manner. The classification models developed in this paper achieve an accuracy rate exceeding 90%, effectively fulfilling the task of classifying images of room layout features. The user’s satisfaction with the interior design methodology experience projects proposed in this paper is also all higher than 70%, and the feedback from both designers and users is very good, which shows the excellent effect of deep convolutional networks in the optimization of interior design visual elements.

In this paper, we use VR equipment to collect relevant facial expression images and normalize the angle, scale, and gray scale of the collected images. The direction quantization of image features is realized by 3D gradient computation, and then the histogram of the direction gradient of each video sub-block is cascaded into the final HOG3D descriptor so as to complete the extraction of dynamic expression features. In view of the multi-dimensional problem of the features, it is proposed to use a principal component analysis algorithm to reduce their dimensionality and use a multi-layer perceptron and deep confidence network to jointly construct the facial expression tracking recognition model. The datasets are used to analyze real-time facial expression tracking in virtual reality. The results present that the verification correctness of both datasets A and B reaches the maximum at the 120th iteration. In contrast, the loss value reaches the equilibrium state quickly at the 40th iteration. The dynamic occlusion expression recognition rate of the deep confidence network on dataset A (66.52%) is higher than that of the CNN (62.74%), which fully demonstrates that the method of this paper is able to effectively improve the performance of real-time facial expression tracking performance in virtual reality. This study can help computers further understand human emotions through facial expressions, which is of great significance to the development of the human-computer interaction field.

In the era of information technology, with the deepening of cultural heritage protection, the importance of digital technology intervention in traditional art and culture for protection and inheritance has become more significant, and the digital inheritance of Chinese traditional music culture has also received widespread attention. In this paper, we first preprocess the traditional music signal and extract the Meier multifrequency coefficients, then use the feature as the input of the deep confidence network, and then keep tuning the parameter training to get the optimal network model, and finally utilize the obtained optimal model to recognize the traditional music. Based on this basis, this paper focuses on the concept of the digital heritage of Chinese traditional music culture through three steps: recognition, processing, display, and dissemination. We also designed and developed a Chinese traditional music culture APP for Chinese traditional music lovers and inheritors and examined the effect and performance of the APP application based on the music recognition model in this paper. The modeling methods proposed in this paper meet the usage requirements by having accuracy rates above 90%. 73.25% of the audience expressed satisfaction and very satisfaction with the overall application effect of the Chinese traditional music APP. In this paper, the Chinese traditional music digital APP proposes an effective means of communication for the solution of digital inheritance and also provides a realizable path for the digital inheritance of Chinese traditional music culture.

At present, in the assembly building construction practice, the relevant subjects lack the concept of building quality management and awareness of responsibility, and the rights and duties of the subjects of each link are not clear in the division of responsibility for building quality. In this paper, based on the deep confidence network, for the problem that the accuracy of the traditional DBN model will gradually decrease, a genetic algorithm is introduced to optimize the conventional restricted Morzmann machine, and the number of nodes in the hidden layer of the genetic algorithm optimization DBN node number process is obtained after the improvement. The optimal method for planning building quality assessment is selected based on the comparison results of the established DBN algorithm function. Then, the optimal building quality assessment model is constructed. Then design the evaluation index system for quality influencing factors and verify it with structural equations. Finally, the model is used to quantify the degree of influence of assembly building quality. The study concludes that all path coefficients affecting the quality of assembled buildings are greater than 0.5, the P-value is less than 0.001, and the five proposed hypotheses are all valid. In the assessment of the quality of residential projects, the final results of excellent, good, moderate, and qualified accounted for 0.1952, 0.2299, 0.3086, and 0.2663, respectively, and the quality of the project’s construction was evaluated as a good grade. This study provides a new method for improving the awareness of quality responsibility among relevant subjects in the construction industry and guaranteeing the level of building quality.

Higher vocational vocal music teaching not only lies in enhancing students’ grasp of notes but also in deep emotional expression and emotional transmission. How to realize the precise classification of emotions expressed by musical style and features is an urgent problem to be solved. In this paper, starting from the classification of acoustic features and music styles, the STFT algorithm is used to extract the music features, and the music emotion model is constructed by combining the two-dimensional emotional psychological model. Based on the Turkish music emotion dataset, the acoustic data are preprocessed by using methods such as frame splitting, pre-emphasis, windowing, and mute frames, a deep confidence network extracts the music features, and the SVM model is introduced to establish the emotion classification model of music style and expression features. To verify the effectiveness of the model, this paper uses the dataset to analyze the music style and emotion classification dimensions, respectively. It is found that the classification accuracy of the Renaissance music category in classical music reaches 97.94%, and the accuracy of the music emotion classification of the model when the music data preprocessing slice granularity is 5 s is 0.838. The DBN-SVM model can realize the emotion classification of the music style and expression characteristics, which can provide a reference for the teachers to carry out targeted vocal music teaching and for the students to realize the release of emotion and emotional resonance when singing in vocal music. It provides a reference for teachers to carry out targeted vocal music teaching and for students to realize emotional release and resonance during vocal singing.

In this paper, in the process of music teaching, the music notes characterized in the time domain are Fourier transformed to the frequency domain, and the resulting notes in the frequency domain are analyzed and processed to obtain the inverted spectral domain features of the notes. On the basis of the cepstrum features, the music recognition model based on the depth confidence network is constructed, and after the training of the depth confidence network, the overfitting phenomenon often occurs for the depth confidence network, and the optimization is carried out by embedding the Dropout method between the implicit layers. From the perspective of music recognition and informationized music teaching, the research on traditional music culture integration of informationized music teaching is designed, and statistical analysis and simulation analysis methods are constructed. The results show that better recognition performance than using music features from different convolutional layers can be obtained using Deep Confidence Networks, with values of 77.5%, 78.8%, and 78.2%, respectively, and that music recognition research based on Deep Confidence Networks is able to better explore and pass on traditional music culture. In the linear regression analysis between the factors of incorporating folk songs into music teaching, the Sig. F and Sig values are 0, which are smaller than the significance level of 0.01 and 0.05, indicating that there is a significant relationship between the factors of students’ gender, age, whether they like folk songs, and the channels of exposure to folk songs and whether folk songs are incorporated into the music teaching program of colleges and universities. This study takes folk songs as representative of traditional music culture, raises people’s awareness of the value of folk songs, and enhances their understanding of the importance of music culture inheritance.

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

At present, deep learning has been widely used many fields, but the research on text classification is still relatively few. This paper makes full use of the good learning characteristics of deep learning, proposes a hybrid model based on deep learning, and designs a text classifier based on the hybrid model. This hybrid model uses two common deep learning models, sparse automatic encoder and deep confidence network, to mix. The hybrid model is mainly composed of three parts, the first two layers are constructed by sparse automatic encoder, the middle layer is a three-layer depth Convolutional Neural Network (CNN), and finally Softmax regression is used as the classification layer. In order to test the classification performance of the classifier based on deep learning hybrid model, relevant experiments were conducted on English data set 20Newsgroup and Chinese data set Fudan University Chinese Corpus. In the English text classification experiment, the classifier based on deep learning hybrid model is used to classify, and a high classification accuracy rate is obtained. In order to further verify the superiority of its performance, a comparative experiment with naive Bayes classifier, K-Nearest Neighbor (KNN) classifier and Support Vector Machine (SVM) classifier demonstrates that the classification effect of the classifier based on deep learning hybrid model is better than that of naive Bayes classifier, KNN classifier and support vector machine classifier. In the experiment of Chinese text classification, the Chinese corpus of Fudan University is tested, and a good classification effect is obtained. The influence of different parameter settings on the classification accuracy is discussed. [ABSTRACT FROM AUTHOR]

Aiming at the problem of high fluctuation and instability of photovoltaic power, a photovoltaic power prediction method combining two techniques has been proposed in this study. In this method, the fast correlation filtering algorithm has been used to extract the meteorological features having a strong correlation with photovoltaic power generation. The complete ensemble empirical mode decomposition with an adaptive noise model has been used to decompose the data into high and low-frequency components to reduce the data volatility. Then, the long short-term neural network and the deep confidence network were combined into a new prediction model to predict each component. Finally, the proposed combined photovoltaic power prediction method has been analyzed using an example and compared with the other prediction methods. The results show that the proposed combined prediction method has higher prediction accuracy. [ABSTRACT FROM AUTHOR]

Given the insufficient early warning capacity of nuclear cold source biological disasters, this paper explores prediction methods for biomass caused by nuclear cold source disasters based on deep learning. This paper also uses the correlation analysis method to determine the main environmental factors. The adaptive particle swarm optimization method was used to optimize the depth confidence network model of the Gaussian continuous constrained Boltzmann machine (APSO-CRBM-DBN). To train the model, the marine environmental factors were used as the main input factors and the biomass after a period of time was used as the output for training. Optimal prediction results were obtained, and thus, the prediction model of biomass caused by the nuclear cold source disaster was established. The model provides an accurate scientific basis for the early warning of cold source disasters in nuclear power plants and has important practical significance for solving the problem of biological blockage at the inlet of cold source water in nuclear power plants.

In order to solve the problem that the evaluation algorithm is easy to fall into local extremum, which leads to slow convergence speed, a skilled talent quality evaluation algorithm based on a deep belief network model was designed. Establish an evaluation set with 4 first level indicators and 14 second level indicators, and calculate the corresponding weights to complete the construction of the evaluation index system. A DBN structure composed of several RBMs and a BP network is constructed. Based on the DBN, a quality evaluation algorithm is designed. The algorithm training is used to evaluate the test data and output the evaluation level. The experimental results show that the convergence speed of DBN based evaluation algorithm is significantly better than that of BP neural network and SVM based evaluation algorithm under the same number of iterations, which is suitable for the accurate evaluation of talent quality. [ABSTRACT FROM AUTHOR]

Copyright of Electric Power Automation Equipment / Dianli Zidonghua Shebei is the property of Electric Power Automation Equipment Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Aiming at the serious threat of heavy haze pollution with PM2.5 as the main pollutant to the national economy and residents' health, an improved PSO combined with deep confidence network was proposed to predict the thickness of atmospheric pollutant PM2.5. Firstly, an improved particle swarm optimization (PSO) fuzzy neural network (FNN) was adopted to integrate algorithm with fuzzy-neural-network, and the characteristics of PSO algorithm in global optimization were given play to forecast the change law of PM2.5 particle thickness. Secondly, a DBN classifier with nonlinear feature extraction preprocessing mechanism is constructed by taking the extracted nonlinear features as input of deep brief networks (DBN). Finally, the effectiveness of the proposed method of predicting PM2.5 thickness of atmospheric pollutants with improved PSO combined with deep confidence network was verified by an example of influencing factor diagnosis and PM2.5 thickness prediction. And the results were compared with the traditional classifier, which showing the advantages of the proposed method in modeling accuracy and convergence speed. [ABSTRACT FROM AUTHOR]

In order to give full play to the value of mental health education courses in colleges and universities and help students understand the relevant knowledge, thus helping to alleviate their own psychological problems, this paper uses the evaluation feedback of mental health courses in colleges and universities to conduct in-depth learning, and further optimize the mental health education courses in colleges and universities. College students, as a seemingly relaxed group, actually bear tremendous pressure. Many pressures, such as study, life, emotion, interpersonal communication, and graduate employment, have seriously affected the mental health of college students. Increasingly prominent mental health problems have attracted the attention of colleges and universities. Domestic colleges and universities have gradually opened mental health education courses, aiming at helping students find breakthroughs in the problem, thereby promoting students' physical and mental development and adapting to the current and developing social environment with the positive and normal psychological state. However, the existing mental health education curriculum in colleges and universities cannot play its role effectively. The general evaluation criteria make the teaching activities less effective, and the students fail to understand the content of the mental health education curriculum well. Therefore, this paper uses the evaluation feedback information of the mental health curriculum in colleges and universities to conduct an in-depth study, optimize the mental health curriculum, and improve the mental health of colleges and universities. The teaching mode of health course enables students to have a deeper understanding of the content of mental health education. It can really help students alleviate psychological pressure and solve psychological problems, which is conducive to the healthy development of universities and society. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Medical Equipment Journal is the property of Chinese Medical Equipment Journal Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Shenzhen University Science & Engineering is the property of Editorial Department of Journal of Shenzhen University Science & Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to understand the environmental trend of network finance and the impact of network finance on traditional ommercial banks, after analyzing the transaction risks of network finance, this study learns from the advanced models of credit risk measurement and early warning at home and abroad and the latest artificial intelligence technology, combining them with China's national conditions, so as to establish a credit risk measurement system suitable for commercial banks. In this study, the research is carried out from the perspectives of theory and practice, technology and business, and the credit risk warning system, measurement model and implementation tools are comprehensively sorted out, then the basic theories and core ideas are studied. In accordance with the concept of big data mining, this study proposes a financial crisis early warning model based on artificial intelligence system. Based on the empirical analysis of the main causes of the rapid decline in the asset quality of China's joint-stock​ commercial banks, and based on the characteristics of big data mining in the information explosion era, it is pointed out that the artificial intelligence is a powerful tool to improve the ability of credit risk measurement. It can be confirmed that the development of network finance has brought different impacts on the business, business model, and business philosophy of banks. [ABSTRACT FROM AUTHOR]