In order to repair the lost data in distributed wind power system, this paper puts forward a wind speed data repairing model based on a new bidirectional prediction method. This model consists of two one-way prediction models. In each prediction model, the original wind speed data are decomposed into several intrinsic mode functions (IMFs) and a residue signal by ensemble empirical mode decomposition (EEMD) method. Then the Savitzky-Golay (SG) filter is used to reduce noise for high-frequency IMFs. Next the long short-term memory (LSTM) model and autoregressive integrated moving average (ARIMA) model are combined to predict low-frequency IMFs and the noise reduction results respectively. At the end, all those forecast results are added and form a one-way result. By weighted average of two one -way results, the repairing result is calculated. The experimental results from multiple prediction cases show that this method can get more accurate results.
Bankruptcy prediction has long been a significant issue in finance and management science, which attracts the attention of researchers and practitioners. With the great development of modern information technology, it has evolved into using machine learning or deep learning algorithms to do the prediction, from the initial analysis of financial statements. In this paper, we will review the machine learning or deep learning models used in bankruptcy prediction, including the classical machine learning models such as Multivariant Discriminant Analysis (MDA), Logistic Regression (LR), Ensemble method, Neural Networks (NN) and Support Vector Machines (SVM), and major deep learning methods such as Deep Belief Network (DBN) and Convolutional Neural Network (CNN). In each model, the specific process of experiment and characteristics will be summarized through analyzing some typical articles. Finally, possible innovative changes of bankruptcy prediction and its future trends will be discussed.
This letter considers the mean-square exponential consensus problem for a class of leader-following multi-agent systems with variable topological structures and nonlinear dynamics. The dynamic of each node in the considered multi-agent system is described by a nonlinear differential equation. The agents transmit signals mutually through the common communication channels, which include time delays, packet dropout, and stochastic disturbances. The variable topological structures are modeled by a Markovian chain. Considering the conditions of the communication networks, a network-based leader-following consensus protocol is proposed. By using Lyapunov functional method and some properties of Kronecker product, some criteria are proposed to ensure the considered multi-agent system achieves H ∞ exponential consensus in the mean square sense. The controller gain matrices could be obtained by solving a minimization problem through a cone complementary technique. The corresponding algorithm is provided.
The electroencephalograph (EEG) microstate is a method used to describe the characteristics of the EEG signal through the brain scalp electrode potential's spatial distribution; as such, it reflects the changes in the brain's functional state. The EEGs of 13 elite archers from China's national archery team and 13 expert archers from China's provincial archery team were recorded under the alpha rhythm during the resting state (with closed eyes) and during archery aiming. By analyzing the differences between the EEG microstate parameters and the correlation between these parameters with archery performance, as well as by combining our findings through standardized low-resolution brain electromagnetic tomography source analysis (sLORETA), we explored the changes in the neural activity of professional archers of different levels, under different states. The results of the resting state study demonstrated that the duration, occurrence, and coverage in microstate D of elite archers were significantly higher than those of expert archers and that their other microstates had the greatest probability of transferring to microstate D. During the archery aiming state, the average transition probability of the other microstates transferring to microstate in the left temporal region was the highest observed in the two groups of archers. Moreover, there was a significant negative correlation between the duration and coverage of microstates in the frontal region of elite archers and their archery performance. Our findings indicate that elite archers are more active in the dorsal attention system and demonstrate a higher neural efficiency during the resting state. When aiming, professional archers experience an activation of brain regions associated with archery by suppressing brain regions unrelated to archery tasks. These findings provide a novel theoretical basis for the study of EEG microstate dynamics in archery and related cognitive motor tasks, particularly from the perspective of the subject's mental state.
Neurofeedback training (NFT) is a non-invasive, safe, and effective method of regulating the nerve state of the brain. Presently, NFT is widely used to prevent and rehabilitate brain diseases and improve an individual’s external performance. Among the various NFT methods, NFT to improve sport performance (SP-NFT) has become an important research and application focus worldwide. Several studies have shown that the method is effective in improving brain function and motor control performance. However, appropriate reviews and prospective directions for this technology are lacking. This paper proposes an SP-NFT classification method based on user experience, classifies and discusses various SP-NFT research schemes reported in the existing literature, and reviews the technical principles, application scenarios, and usage characteristics of different SP-NFT schemes. Several key issues in SP-NFT development, including the factors involved in neural mechanisms, scheme selection, learning basis, and experimental implementation, are discussed. Finally, directions for the future development of SP-NFT, including SP-NFT based on other electroencephalograph characteristics, SP-NFT integrated with other technologies, and SP-NFT commercialization, are suggested. These discussions are expected to provide some valuable ideas to researchers in related fields.
The COVID-19 epidemic has caused great impact on the entire society, and the spread of novel coronavirus has brought a lot of inconvenience to the education industry. To ensure the sustainability of education, distance education plays a significant role. During the process of distance education, it is necessary to examine the learning situation of students. This study proposes an academic early warning model based on long- and short-term memory (LSTM), which firstly extracts and classifies students’ behavior data, and then uses the optimized LSTM to establish an academic early warning model. The precision rate of the optimized LSTM algorithm is 0.929, the recall rate is 0.917 and the F value is 0.923, showing a higher degree of convergence than the basic LSTM algorithm. In the actual case analysis, the accuracy rate of the academic early warning system is 92.5%. The LSTM neural network shows high performance after parameter optimization, and the academic early warning model based on LSTM also has high accuracy in the actual case analysis, which proves the feasibility of the established academic early warning model.
Objective: The goal of this article is to study the relationship between innovations and exporting of Chinese firms and identify which type of innovation contributes most to the probability of exporting. Research Design & Methods:We refer to the recent strand in the new trade theory literature that stresses the importance of firm productivity in entering export markets.We distinguish between product, process and managerial innovations that can increase productivity.The empirical investigation is based on the probit model and the firm-level data set covering two years: 2003 and 2012.Findings: Our empirical results show that the probability of exporting is positively related to product and process innovations, firm size, foreign capital participation and foreign technology.Moreover, we find that in 2003 process innovations were more important for export performance than product innovations, while in 2012 it was the opposite.Implications & Recommendations: Firms should coordinate their strategic assets and resources for innovation in order to enhance their overall level of competitiveness.Governments should work on establishing stronger institutional environment necessary to provide firms with protection of intellectual property rights, an easier access to financing of innovation, a lower tax burden upon innovative firms, the higher quality of human resources to firms and more supportive policy packages. Contribution & Value Added:In contrast to previous that used only the R&D spending as the measure of innovation in our study we also use innovation outcomes.In particular, we determine which innovation type is of greatest importance, having controlled for the set of other firm characteristics.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)