Search

Aiming at the problem of traffic data distribution shift caused by new unknown attacks in the industrial Internet, a malicious traffic identification method based on neighborhood filtering and stable learning was proposed to enhance the effectiveness and robustness of the existing graph neural network model in identifying known malicious traffic. Firstly, the graph structure of the traffic data was modeled to capture the topological relationship and interaction mode in communication behavior. Secondly, the traffic subgraph was divided based on the neighborhood filtering mechanism of biased sampling to eliminate the pseudo-homogeneity between communication behaviors. Finally, the statistical independence of high-dimensional traffic features was realized by applying graph representation learning and stable learning strategies, combined with adaptive sample weighting and collaborative loss optimization methods. The experimental results on two benchmark datasets show that compared with the baseline method, the recognition performance of the proposed method is increased by more than 2.7% in the new unknown attack scenario, which shows its high efficiency and practicability in the industrial Internet environment.

Land consolidation has become an organized and widely implemented project in rural areas. However, research remains somewhat scarce on the contributing mechanism of land consolidation to farmer income in developing countries. Using provincial panel data from China, this study employs a fixed effects model to analyze the influence of high-standard farmland construction on farmer income and underlying mechanisms. Results suggest that the implementation of high-standard farmland construction can significantly augment farmer income. Reduction in agricultural production cost, improvement in agricultural management benefit and increase in non-farm income have led to the boost in income. The relationship between income and high-standard farmland construction exhibits regional disparities, with the most significant impacts concentrated in economically developed areas and low relief degree areas. Our findings suggest that it is imperative for the Chinese government to persistently promote the establishment of high-standard farmland, and further enhance its positive influence on advancing agricultural cost-effectiveness and expanding non-farm income channels for farmers.

Fault-fracture bodies are developed in the southwestern margin of Ordos Basin， and thus， the study on the controlling effect of fracture bodies on sweet spot distribution of Yanchang Formation reservoirs needs to be deepened. In this paper， based on the extensive seismic and logging interpretation results， the characteristics of fault-fracture bodies are described from the point of view of their evolution mechanism， and then the controlling effect of fault-fracture bodies on oil and gas is systematically discussed. The results show that vertical strike-slip faults are developed in the southwest margin of Ordos Basin， with Y-shaped， flower-like， and negative flower-like structures. The faults usually pass through the bottom of Chang 7 Member， the bottom of Yan’an Formation， the bottom of the Cretaceous series， and the basement. Some of the faults retain the early reverse fault properties， indicating that the late inversion degree is not complete. The main fault has different shapes and migrations in different parts， and the section shows the cyclic transformation of tension-torsion and compression-torsion properties. On the plane， different types of fracture combinations appear alternately. The development model of strike-slip faults is established， and the strike-slip faults have typical characteristics of “multi-stage activity and inherited development”. Chang 8 Member mainly develops vertical fractures and horizontal bedding fractures. The development frequency of horizontal bedding fractures is 62.5%， while that of vertical fractures is 37.5%. The oil level of vertical fractures is relatively higher. Fractures are mainly developed in fine sandstone in distributary channels. When the distance from the main fault is greater than 1.25-1.5 km， the degree of fracture development decreases sharply， showing a fault-fracture body boundary. In addition， fractures are relatively developed in the single sand body within 6 m from the main fault. When the thickness of the single sand body exceeds 6 m， the degree of fracture development decreases sharply. The study shows that the index system based on sedimentation （foundation）， structure （dominant）， and fracture （effective） can effectively predict the favorable zones of Chang 8 fault-fracture body reservoir.

This paper investigates the impact of digital transformation on innovation within the agribusiness sector, both theoretically and empirically, through an examination of Chinese A-share agriculture-related listed companies spanning from 2011 to 2021. The findings suggest that digital transformation significantly enhances innovation capability of agribusiness, while concurrently fostering improvements of its innovation quality. These results hold robust following an endogeneity test and a series of robustness tests. Our heterogeneity analysis found that digital transformation exerts a more pronounced influence on promoting innovation among state-owned agribusinesses, those located in the eastern region of China, and those facing heightened financing constraints. Mechanism testing revealed that digital transformation not only enhances the technological capabilities of agribusiness but also alleviates their financial constraints, thereby facilitating the convergence of innovative resources such as technology, talent and capital to agribusiness, and consequently elevating agribusi-ness innovation levels. This paper elucidates the impacts and mechanisms of digital transformation on agribusiness innovation, offering valuable insights for decision-makers aiming to foster agribusiness innovation, particularly in developing nations.

Objective: In order to solve the problem of manual weighting of dragon fruit, including time-consuming, laborious and expensive, an automated weight estimation method based on machine vision and machine learning was proposed in this research. Methods: Firstly, 106 dragon fruits were weighed, recorded and photographed, and images of dragon fruits were constructed. Secondly, binary images were obtained after denoising and segmentation. Moreover, the three features of pixel area, major axis pixel length and minor axis pixel length of dragon fruits were extracted on the basis of binary images. The three features of each image and their corresponding weights were combined into a set of data, which was divided into training set and test set according to the ratio of 7∶3. Finally, the training set was input into the Gradient Boosting, Random Forest, K-Neighbors and Artificial Neural Networks machine-learning models for training, and the test sets were used for model evaluation. Results: The evaluation index of the Artificial Neural network performed well compared with other models, with R2 of 0.986 and RMSE of 13.091. Conclusion: The experimental result demonstrates that the method proposed in this research can accomplish the weight estimation of dragon fruit effectively, and meet the weight estimation requirements of dragon fruit.

Objective: An analytical method was established for determination of 13 kinds of bisphenols and alkylphenol compounds in liquid dairy products by salting-out-assisted liquid-liquid extraction coupled with HPLC-MS/MS. Methods: The salting-out-assisted liquid-liquid extraction pre-treatment method was adopted, and the detection was performed in the mode of negative ion source multiple reaction monitoring （MRM） by triple quadrupole mass spectrometry, and quantitative analysis was performed by matrix external standard method. Besides, the effect of extraction solvent type, selection and the usage of salting-out agents on the recoveries were studied and the performance of the method was evaluated. Results: The liquid dairy products were liquid-liquid extracted with acetonitrile as extraction solvent and 0.4 g sodium chloride as salting-out agents. The 13 kinds of phenols had a good linear relationship, and the correlation coefficients were all ＞ 0.99. The standard recovery rates of milk and yoghurt were 81.5% to 118.1% and 80.8% to 107.6%, respectively, and the relative standard deviations were 1.89% to 12.3% and 1.12% to 11.9%. The detection limits were 0.15 to 0.75 μg/kg, and the quantification limits were 0.5 to 2.5 μg/kg. One of the commercially available liquid dairy product was found to contain 4-NP, and the concentrations was 9.28 μg/kg. Conclusion: The established method is suitable for the rapid detection of 13 kinds of bisphenols and alkylphenol compounds in liquid dairy products.

Nowadays, the Muslim community is under the spotlight. The ongoing international debate over Muslim minorities identities reflects the dilemma of their integration into non-Muslim modern multi-ethnic nations under the context of multiculturalism. One typical example is the heated debate between the West and China over Chinese Muslims, also known as Uyghurs, through media discourse. Western media, such as the BBC, claim that China assimilates and forces Uyghurs to integrate. The Chinese side, meanwhile, asserts that mainstream Western media reports are lies, smears, and fake. Facing the discourse challenges from the West, China applies media discourse to reinforce its understanding of Uyghurs in the international discourse system. The analysis reveals that the Chinese government has consciously aligned the image of Uyghurs with its Chinese national image, and building a real Xinjiang image on CGTN.

The quality of cultivated land is an important basis for guaranteeing grain yield, and improving the quality of cultivated land is an important initiative of the Chinese government to implement its food security strategy. This paper explores the effects of formal technical training and informal online self-directed learning on farmers’ willingness to adopt cultivated land conservation technology using farmer-level survey data. The results show that farmers’ participation in technical training effectively increased their willingness to adopt straw return technologies, but participation in online self-directed learning did not affect farmers’ willingness to adopt, while farmers who participated in both technical training and online self-directed learning had stronger willingness to adopt. These results show that internet-based self-directed learning is a useful supplement to the formal training system. Further analysis revealed that participation in technical training increased farmers’ awareness of the usefulness of the straw-return technology, which in turn increased their willingness to adopt the conservation technology of cropland. This paper clarifies the impact and mechanisms of the two main existing technology learning modes on farmers’ adoption of new technologies in rural China and provides a reference for the promotion of agricultural technology diffusion and the improvement of the quality of cultivated land in China.

Introducing defects in photonic crystals is a common method for manipulating and controlling the propagation of electromagnetic waves. By introducing defects in photonic crystal waveguides, the periodicity of the waveguide structure can be disrupted, local modes can be formed, and resonant cavity functions can be achieved. In this study, we designed two groups of two-dimensional valley photonic crystal waveguides, each of which uses different methods to introduce defects and obtained different resonant cavity structures, and designed a resonant cavity sensor. We conducted a detailed theoretical analysis of the resonant cavity through simulation software. In addition, we fabricated the samples and conducted microwave experiments to demonstrate the accuracy of our theoretical research. Our research provides guidance for the application of photonic crystal devices. [ABSTRACT FROM AUTHOR]

Photoplethysmography (PPG) signal is potentially suitable in atrial fibrillation (AF) detection for its convenience in use and similarity in physiological origin to electrocardiogram (ECG). There are a few preceding studies that have shown the possibility of using the peak-to-peak interval of the PPG signal (PPIp) in AF detection. However, as a generalized model, the accuracy of an AF detector should be pursued on the one hand; on the other hand, its generalizability should be paid attention to in view of the individual differences in PPG manifestation of even the same arrhythmia and the existence of sub-types. Moreover, a binary classifier for atrial fibrillation and normal sinus rhythm is not convincing enough for the similarity between AF and ectopic beats. In this study, we project the atrial fibrillation detection as a multiple-class classification and try to propose a training pipeline that is advantageous both to the accuracy and generalizability of the classifier by designing and determining the configurable options of the pipeline, in terms of input format, deep learning model (with hyperparameter optimization), and scheme of transfer learning. With a rigorous comparison of the possible combinations of the configurable components in the pipeline, we confirmed that first-order difference of heartbeat sequence as the input format, a 2-layer CNN-1-layer Transformer hybridR model as the learning model and the whole model fine-tuning as the implementing scheme of transfer learning is the best combination for the pipeline (F1 value: 0.80, overall accuracy: 0.87)R.