Table of Content

15 June 2026, Volume 11 Issue 6

Previous Issue   

Global Error Bound Estimation for the Extended Vertical Linear Complementarity Problem of S-SDD Matrix and SB-Matrix

Wang Feng, Zhou Jiabing

2026, 11(6):  1-10. 

Asbtract ( 0 )   PDF (1313KB) ( 0 )  

References | Related Articles | Metrics

The extended vertical linear complementarity problem is a generalization of the linear complementarity problem and has important applications in fields such as engineering, economics, and control theory. By leveraging the upper bounds of the infinity norms of the inverse of S-SDD matrices and SB-matrices, combined with inequality amplification techniques, new estimation formulas for the global error bound of the extended vertical linear complementarity problems of S-SDD matrices and SB-matrices are derived, improving upon existing results. Numerical examples validate the effectiveness of the new estimation formulas.


Several Studies on Complete t-Partite Graphs

Yang Huang, Yang Zhongyan, Yang Limin

2026, 11(6):  11-18. 

Asbtract ( 0 )   PDF (1250KB) ( 0 )  

References | Related Articles | Metrics

The article presents several achievements, problems, and conjectures in the study of complete t-partite graphs from seven
perspectives: (1) the counting formulas of S(n)-factors; (2) the counting formulas of the partitions of stable sets; (3) chromatic polynomials and unimodality; (4) mean color numbers; (5) independence numbers and independence polynomials; (6) the Hosoya index and the Merrifield-Simmons index; (7) the chromatic uniqueness. These findings enrich the intersection of extremal graph theory and algebraic graph theory.


Design and Implementation of a Federated Learning Framework with End-Edge-Cloud Collaboration

Yang Yuelang, Wang Xinchen, Yang Yiyu, Yang Dengqi, Li Xiaowei

2026, 11(6):  19-26. 

Asbtract ( 0 )   PDF (1634KB) ( 0 )  

References | Related Articles | Metrics

With the increasing demand for data privacy protection, federated learning has emerged as a crucial method for safeguarding data privacy due to its ability to enable joint model training without sharing raw data. However, federated learning still faces numerous challenges in practical applications, such as poisoning attacks by malicious clients, high communication costs, and excessive computational load on central servers. To address these issues, a federated learning framework with end-edge-cloud collaboration is proposed. By hierarchically managing clients, edge servers, and cloud servers, this framework effectively reduces the communication burden on cloud servers while improving system scalability and efficiency. Additionally, a federated learning algorithm was designed
within this framework, combining reputation-based local model aggregation and fully homomorphic encryption-based global model ag⁃
gregation to effectively counteract data poisoning attacks. Experimental results demonstrate that the federated learning framework with end-edge-cloud collaboration not only enhances the system's defense against malicious attacks but also significantly reduces communication overhead and improves system efficiency.


EDG-DQN: An Enhanced Hybrid Architecture for Influence Maximization via Deep Reinforcement Learning

Li Yuxiang, Zhang Jinming, Zhang Jun, Ruan Huiqiong, Luo Guilan

2026, 11(6):  27-38. 

Asbtract ( 0 )   PDF (2662KB) ( 0 )  

References | Related Articles | Metrics

To address the issue of degraded propagation performance of seed node sets in large-scale social networks by existing deep reinforcement learning models, this study proposes an end-to-end deep reinforcement learning model called EDG-DQN. First, it em⁃
ploys an attention mechanism to focus on key information, constructing a dual-channel graph attention network to learn embedded representations of nodes in both the influence application space and the influence reception space. Second, a variant of the long short-term memory network structure——the dynamic influence unit is introduced, treating the environmental state in reinforcement learning as a temporal feature to dynamically enhance the spatial embedded representations of nodes. Finally, the deep Q-network approximates the Q-function to achieve an optimal selection strategy for nodes. Comparative experiments based on dynamic simulated network propagation and simulated real-world network computational costs demonstrate that the seed node sets selected by this model on large-scale sparse social network datasets exhibit superior influence and diffusion efficiency compared to existing benchmark models.


Data Classification Method for College Students' Innovation and Entrepreneurship Projects in Applied Higher Vocational Colleges

Huang Xuefei, Gong Xiaohui

2026, 11(6):  39-45. 

Asbtract ( 0 )   PDF (1649KB) ( 0 )  

References | Related Articles | Metrics

Data classification is a crucial component in the management of innovation and entrepreneurship project data for vocational college students. However, due to data imbalance, classifiers may suffer from insufficient generalization capabilities, leading to less accurate classification results. To address this issue, this study proposes a data classification method for innovation and entrepreneurship proj⁃
ects. First, textual data from the project database is converted into vector form, and a machine learning classifier based on support vector machines is established. Then, an adaptive synthetic sampling method is employed to oversample the classifier, mitigating data imbalance and optimizing its generalization capabilities. Finally, the transformed project data is classified using the classifier, and the K-means algorithm is applied for clustering integration to achieve project data classification. Experimental results demonstrate that this method achieves high accuracy in classifying innovation and entrepreneurship data, with a mean average precision of 0.986 and an F1 score of 0.973, exhibiting excellent classification performance and promising practical application prospects.


Research on Multi-Time Scale Scheduling Optimization of Microgrids Based on IASO

Zhang Jiawei, Hu Yongmao, Sun Xiugui

2026, 11(6):  46-53. 

Asbtract ( 0 )   PDF (1686KB) ( 0 )  

References | Related Articles | Metrics

The grid-connected microgrid encompasses photovoltaic cells, wind turbines, battery energy storage systems, diesel generators, and variable loads. The complex coupling relationships among its internal equipment, coupled with the inherent uncertainty of wind and solar renewable energy and the volatility of loads, pose challenges to microgrid dispatching. To address this, this study constructs an optimized microgrid model integrated with a demand response mechanism for electricity prices, with the core objective of minimizing the overall system operating cost. Through an improved atom search optimization, multi-time scale economic optimal dispatching is achieved. To ensure the economic feasibility and practicality of the strategy, the study employs an optimized atom search optimization to solve for the economic optimum of the microgrid model across various time scales. This initiative aims to enable rapid response to internal fluctuations in the microgrid system while ensuring optimal overall economic performance.



Diversity of Autumn Banding Birds in Yunnan Weishan Niaodaoxiongguan from 2019 to 2023

Zhang Guoli, Cui Maohuan, Liu Lanxiang, Liu Xiaoting, Guan Xingtian, Huang Jiankun , Yang Xun

2026, 11(6):  54-67. 

Asbtract ( 0 )   PDF (1531KB) ( 0 )  

References | Related Articles | Metrics

An analysis of the nocturnal migratory bird banding data from the Weishan Niaodaoxiongguan Bird Banding Station from 2019 to 2023 revealed the following： (1) Over five years, a total of 13 504 individuals from 181 migratory bird species were banded, belonging to 13 orders and 34 families, with the Ficedula albicilla, Calliope calliope, Lanius cristatus, and Anthus hodgsoni being the domi⁃
nant species; (2) The Shannon-Wiener diversity index for birds was 3.075 2, the Pielou evenness index was 0.587 3, the dominance index was 0.137 3, and the concentration index was 0.561 4; (3) The inter-family diversity index, inter-genus diversity index, and spe⁃
cies diversity G-F index for birds were 17.495 8, 4.122 3, and 0.764 4, respectively; (4) The highest similarity coefficient among
banded birds across different years was 0.750 0, while the lowest was 0.530 0; (5) Thirteen species of nationally protected wild birds
were recorded, including five vulnerable or higher-threatened species listed in the China Red List of Biodiversity, three in the IUCN
Red List of Threatened Species, and three in the appendices of the Convention on International Trade in Endangered Species of Wild
Fauna and Flora. Based on these findings, corresponding conservation management recommendations for migratory birds at Weishan
Niaodaoxiongguan were proposed.


Screening, Identification and Enzymatic Characterization of a High-Yield Urate Oxidase-Producing Strain from Streptomyces

Liu Hongyan , Yin Yirui , Sang Peng

2026, 11(6):  68-74. 

Asbtract ( 0 )   PDF (2203KB) ( 0 )  

References | Related Articles | Metrics

Urate oxidase catalyzes the decomposition of uric acid into hydrogen peroxide and allantoin, which has higher solubility, and is commonly used for uric acid detection and treatment of hyperuricemia-related diseases. In this study, a high-yield urate
oxidase-producing strain YHBJ-12 (GenBank accession number ON351059.1) was isolated from sediment samples of Chaka Salt Lake
in Qinghai Province. Based on 16S rRNA gene sequencing and phylogenetic analysis, the strain was identified as Streptomyces sp. Enzymatic properties revealed that the optimal reaction pH for the urate oxidase produced by strain YHBJ-12 was 7.0, and the optimal reaction temperature was 25 °C. The enzyme maintained over 60% relative activity within a pH range of 5.6-8.6 and a temperature range of 15-50 °C, demonstrating good pH stability and thermal stability. It retained high activity under near-human physiological conditions (37 °C, pH 7.0). K⁺, Mg²⁺, and Co²⁺ significantly enhanced enzyme activity, whereas Ba²⁺, Mn²⁺, Pb²⁺, Cu²⁺, and inhibitors including EDTA, SDS, PMSF, and DTT exerted varying degrees of inhibitory effects. These characteristics indicate that the urate oxidase from strain YHBJ-12 possesses potential for further development and application, providing an excellent enzyme source for the development of high-sensitivity uric acid detection reagents and novel urate oxidase preparations.


Construction and Application of a 3D Virtual Simulation Experiment for the External Identification and Digestive System Anatomy of Dengchuan Cattle, a Yunnan Specialty Breed

Zhao Tianzhang, Zhao Tianzhi, Long Xiaowen, Wu Mingcan, Li Huiying

2026, 11(6):  75-81. 

Asbtract ( 0 )   PDF (2284KB) ( 0 )  

References | Related Articles | Metrics

In response to the endangered status of purebred Dengchuan cattle and the long-term difficulties caused by insufficient teaching resources in large-scale experimental animal anatomy teaching in universities, this project constructs a "3D Virtual Simulation
Experiment Platform for the External Identification and Digestive System Anatomy of Dengchuan Cattle". On the one hand, it realizes
the rescue construction of a purebred Dengchuan cattle digital specimen resource library and provides data support for the construction of the germplasm resource library of Dengchuan cattle, the only dairy yellow cattle breed in China. On the other hand, with the help of new generation information technologies such as virtual simulation, this platform integrates with the practical teaching system of animal science majors, and builds a multidimensional practical teaching system that is in line with regional industrial characteristics. By implementing the experimental teaching mode of "combining virtual and real" and the formative experimental assessment system, on the basis of fully mobilizing students' learning interest and initiative, students' practical ability and experimental teaching effectiveness are significantly improved, effectively avoiding the duplication and waste of teaching resources.


Research on the Coupling Coordination between Regional E-Commerce and Economic Growth

Xu Na

2026, 11(6):  82-92. 

Asbtract ( 0 )   PDF (2419KB) ( 0 )  

References | Related Articles | Metrics

As a new business format and mode, e-commerce can promote steady, efficient, and high-quality economic growth, while its own development also relies on the support of economic foundations. Based on the panel data of 31 provinces in China from 2014 to 2022,this study measures the development levels of e-commerce and economic growth as well as their coupling coordination relationship from multiple dimensions. The results show that： both e-commerce and economic growth in China exhibit a pattern in which eastern coastal provinces outperform those in the central and western regions; the coupling degree between e-commerce and economic growth is relatively high, but the overall coordination level remains moderate, with this issue being particularly pronounced in the western provinces;the spatial disparities in coupling coordination are primarily attributed to inter-regional differences, and the intra-regional disparities within the eastern provinces are the most significant. Additionally, the e-commerce scale, total economic output, and economic efficiency are identified as the core constraints on the coupling coordination relationship between these two systems. Based on these findings, policy recommendations are proposed, including strengthening government support for e-commerce development in the western region and promoting intra-regional collaborative cooperation.


Research on Bus Routes Optimization in Dali City Based on Graph Theory and Genetic Algorithms

Li Wu, Zhang Chaoyuan, Zhou Shaoyan, Li Jiamei

2026, 11(6):  93-100. 

Asbtract ( 0 )   PDF (1687KB) ( 0 )  

References | Related Articles | Metrics

Taking the public transportation system of Dali City as the research subject, this study addresses issues such as prolonged
waiting times and route congestion caused by the surge in passenger flow during tourist peak seasons. By integrating graph theory and genetic algorithms, the research conducts an optimization study on bus routes. First, a network topology structure is constructed with bus stops as nodes and comprehensive travel costs as edge weights. Subsequently, a multi-objective optimization model is established with the goals of minimizing passenger waiting time and operational costs. The genetic algorithm is employed to solve this model, yielding an optimized bus scheduling scheme. MATLAB simulations are then conducted for verification and comparative analysis. The results demonstrate that the optimized scheme effectively enhances bus operational efficiency, providing theoretical support and practical reference for the  optimization of public transportation systems in tourist cities.