EAI Endorsed Transactions on Scalable Information Systems

The Quality Evaluation of Innovation and Entrepreneurship Education in Colleges and Universities in the Context of Big Data

Xiaoxue Fan — 2025-05-14

INTRODUCTION: The advancement of appropriate instructional procedures and data techniques had already facilitated the development of a virtual evolutional innovation and entrepreneurship education process.

OBJECTIVES: The adoption of big data has also tended to result in academic revolutionary movements for businesses. To address the limitations of current university instructional setups and to broaden the scope of Big Data's (BDC) application, this study presents an optimization technique for evaluating the quality of innovation and entrepreneurship education.

METHODS: This paper presents an optimization technique for measuring the quality of education in innovation and entrepreneurship. This technique should help us move beyond the constraints of our existing approaches to higher education's pedagogical infrastructure into the domain of big data.

RESULTS: The proposed optimization algorithm differs from conventional university quality evaluation instructional practices as it integrates big data that dramatically improves the college Entrepreneurship class interaction. This article proposed a Convolution neural network algorithm with BDC for training virtualization representation and employs a standard correlation analysis method in data analysis to retrieve the correlation relationship between the information content enclosed in huge entrepreneurship education and online students.

CONCLUSION: Simulation results revealed that the proposed model has provided an accuracy of 98%. The proposed method provides a platform for sharing instructional materials that function more efficiently under heavy load. Up to 98% and 94% security are achieved under heavy and light loads, respectively. The use of cloud computing in this scenario led to improvements of 7% and 8%, respectively, yielding results of 89% and 86%.

Fuzzy TOPSIS Method for Sustainable Supplier Assortment in Green Supply Chain Management

Huzaifa Saleem — 2025-05-26

INTRODUCTION: Green supply chain management represents one of the crucial ways for organizations to start minimizing their ecological footprint in the era of increasing ecological preoccupation and sustainability objectives. The critical issues of green supply chain management involve the assortment of green suppliers who are well-suited with the environmental objectives of organizations. Traditional methods of supplier selection cannot efficiently depict the complex, uncertain nature of sustainability criteria. In this respect, the Fuzzy Technique for Order of Preference by Similarity to Ideal Solution, shortly known as Fuzzy TOPSIS, is proposed for usage in this research for improving the process of supplier assortment in green supply chain management.

OBJECTIVES: The aim of this work is, therefore, to present an integrated framework, utilizing the Fuzzy TOPSIS method for selecting sustainable suppliers in green supply chain management. The particular aim of the study will be to incorporate environmental, social, as well as economic criteria in performance evaluation at the supplier level, by considering innate uncertainties and fuzziness related to sustainability metrics.

METHODS: The Fuzzy TOPSIS process is applied to assess and rank potential suppliers based on multiple criteria considering both environmental and economic factors.

RESULTS: Application of the Fuzzy TOPSIS method in sustainable supplier assortment demonstrates its effectiveness in identifying suppliers that align with green objectives while meeting operational requirements.

CONCLUSION: The proposed framework will provide a more fine-tuned and flexible tool for decision-makers by incorporating fuzzy logic into the complexities at hand for sustainability assessment. The findings underline the importance of adopting advanced techniques in decision making in order to attain environmental responsibility and long-term sustainability in supply chain operations.

Research on Optimization of Unmanned Aerial Vehicle Communication Based on Wireless Communication Artificial Intelligence

Yu Geng — 2025-07-04

In the last few years, there has been an increasing use of Unmanned Aerial Vehicles (UAVs), leading to a demand for robust and efficient communication technologies. But by their nature, UAVs operate dynamically and relatively sparsely in the sky, making it inconvenient to perform traditional communication. To overcome communication hurdles, the authors of this research propose an AI approach to optimize wireless communications to enhance UAV (Unmanned Aerial Vehicles) performance. The study will investigate the development of an AI-based algorithm to adjust communication parameters concerning various aspects, including UAV movement, environmental conditions, and network constraints. It will increase the data transmission quality, reduce latency, and service larger areas the network covers for UAVs. Simulations and actual world experiments will be conducted to evaluate the proposed approach, showing that AI can enhance UAV communication. This work could significantly improve the performance of UAVs and optimize their application in many fields ranging from surveillance to delivery services.

Research on Hybrid Path Planning Algorithms for UAVs in Complex Environments

Xinyue Chang — 2025-05-27

INTRODUCTION: This paper investigates a UAV path planning algorithm in a UAV-assisted network scenario, integrating both global and local path planning. Firstly, the ASPSO (Adaptive Spherical Vector-Based Particle Swarm Optimization) algorithm is proposed for offline path planning to obtain key global path points, providing a general flight strategy for the UAV. During the flight, the UAV continuously detects surrounding obstacles in real-time. If newly detected obstacles are encountered, the ECAVF (Enhanced Collision Avoidance Vector Field) algorithm is employed for local path planning to dynamically avoid obstacles and ensure the safety of the UAV.

OBJECTIVES: The objective of this paper is to enhance the path planning capability of existing algorithms in complex three-dimensional environments, enabling UAVs to operate efficiently and safely.

METHODS: The proposed ASPSO algorithm determines parameter ranges for different scenarios during the initialization phase, effectively reducing initialization time. Additionally, a multi-strategy optimization approach is introduced during the search process. Expanding the search space in the early iterations helps escape local optima, while minor perturbations are introduced in the later iterations to continue exploring within the neighbourhood of high-quality solutions. Finally, a method utilizing virtual control points for path refinement is proposed to smooth the trajectory. The ECAVF algorithm incorporates a dynamic adjustment factor based on relative velocity to optimize the vector field in the presence of multiple moving obstacles. By integrating factors such as distance and velocity, a hybrid vector field is constructed, demonstrating superior robustness in complex multi-obstacle scenarios.

RESULTS: The proposed method is compared with the PSO (Particle Swarm Optimization), the Spherical Vector-based PSO, and the original CAVF (Collision Avoidance Vector Field) method. The results demonstrate that the proposed method exhibits higher initialization efficiency, superior initial solution quality, and the ability to obtain a more optimal global path. Additionally, it shows stronger dynamic obstacle avoidance capabilities and a higher success rate in avoiding obstacles.

CONCLUSION: These results demonstrate that the proposed method effectively enhances the quality of global path planning solutions and improves the success rate of dynamic obstacle avoidance.

Smart Data Prefetching Using KNN to Improve Hadoop Performance

Rana Ghazali — 2025-04-17

Hadoop is an open-source framework that enables the parallel processing of large data sets across a cluster of machines. It faces several challenges that can lead to poor performance, such as I/O operations, network data transmission, and high data access time. In recent years, researchers have explored prefetching techniques to reduce the data access time as a potential solution to these problems. Nevertheless, several issues must be considered to optimize the prefetching mechanism. These include launching the prefetch at an appropriate time to avoid conflicts with other operations and minimize waiting time, determining the amount of prefetched data to avoid overload and underload, and placing the prefetched data in locations that can be accessed efficiently when required. In this paper, we propose a smart prefetch mechanism that consists of three phases designed to address these issues. First, we enhance the task progress rate to calculate the optimal time for triggering prefetch operations. Next, we utilize K-Nearest Neighbor clustering to identify which data blocks should be prefetched in each round, employing the data locality feature to determine the placement of prefetched data. Our experimental results demonstrate that our proposed smart prefetch mechanism improves job execution time by an average of 28.33% by increasing the rate of local tasks.

Enhancing Stakeholder Analysis with AI: A Comparative Study of Productivity and Quality in the Educational Context

Vijay Kanabar — 2025-05-23

This paper examines the application of generative artificial intelligence in stakeholder management while studying the business aspects of software development and project management in two different universities. It explores a novel intersection of AI with software development and project management practices, offering valuable insights for both academia and industry. By investigating how students use AI alongside traditional methods under supervision, this study evaluates the effectiveness, quality of results, and creativity of students’ project assignments in identifying stakeholders and defining communication strategies. The findings suggest that AI can enhance work completion speed and contribute to greater project success due to a more complete identification of stakeholders and formulation of innovative stakeholder engagement strategies. There is a consensus, within this context, that while AI can be invaluable for project stakeholder management, human judgment remains essential.

On the Performance of RIS-Enhanced RSMA Networks with On-Off Control

Haiyan Huang — 2025-06-11

As a pivotal enabling technology for the 6G wireless communication system, Reconfigurable intelligent surface (RIS) is capable of dynamically adjusting its electromagnetic elements, thereby significantly enhancing wireless network coverage, spectral efficiency, and energy sustainability. In this context, this paper investigates the integration of RIS into the rate-splitting multiple access (RSMA) framework, where the base station employs the RIS to transmit superimposed signals to multiple users simultaneously. To reduce computational complexity, a On-Off control scheme is adopted, considering both imperfect successive interference cancellation (ipSIC) and perfect successive interference cancellation (pSIC) scenarios. closed-form expressions for the exact outage probability of the $k$-th user under Rayleigh fading channels are derived. Furthermore, to more comprehensively evaluate system performance, asymptotic outage probability expressions for ipSIC/pSIC in the high signal-to-noise ratio (SNR) regime are also derived. The results demonstrate that, under identical system configurations, the RSMA-based communication scheme outperforms the non-orthogonal multiple access (NOMA) scheme, exhibiting superior spectral resource utilization and interference management capabilities. In the pSIC scenario, the user diversity order is contingent upon the number of RIS reflecting elements and the channel ordering approach. Increasing the number of RIS reflecting elements can effectively compensate for the performance degradation caused by residual interference in the ipSIC scenario, thereby further enhancing system performance

A Cholesky decomposition and fusion clustering based spectrum sensing method

Jian Li — 2025-06-24

Spectrum sensing is a key technology to detect unused frequency bands, and is widely applied in spectrum sharing and dynamic channel allocation. However, it is a challenge to provide high sensing accuracy under low signal-to-noise-ratio (SNR) environments. To address this issue, this paper proposes a novel method based on feature extraction and fusion clustering. First, the sampling matrix of the received signal is decomposed into two orthogonal components I and Q, and Cholesky decomposition is performed on the covariance matrices of I and Q components to extract their two-dimensional feature vectors. Then, the fusion clustering algorithm is proposed, where the GMM clustering algorithm is performed to classify the feature vectors, and the initial parameters of GMM, such as centroids, weights and covariance matrices, are generated by K-means clustering. Simulation results show that the proposed method accelerates the convergence speed of GMM and improves the classification accuracy. It effectively enhances the performance of spectrum sensing compared to other mainstream methods.

Mobile Edge Computing Empowered Energy Consumption Optimization for Multiuser Power IoT Networks

Shuangwei Li — 2025-06-18

Mobile edge computing (MEC) has emerged as a promising solution to enhance the computational capabilities of resource-constrained IoT devices while optimizing energy consumption. In this paper, we investigate energy-efficient resource allocation strategies for multiuser power IoT networks by jointly optimizing the offloading ratio, transmit power, and wireless bandwidth allocation. We propose an optimization framework that minimizes the total energy consumption of the system while ensuring the latency constraints of computation tasks. To solve this challenging non-convex problem, we employ an alternating optimization approach, where the offloading decision, wireless bandwidth allocation, and transmit power control are iteratively refined using convex optimization techniques and successive convex approximation (SCA). Simulation results are provided to show that the proposed scheme significantly outperforms the competing approaches in terms of energy efficiency. Specifically, for a system with 6 users, the proposed scheme maintains an energy consumption of approximately 0.1 Joules, reducing the energy consumption of the conventional schemes to less that 40 percentages.

Deep Learning Empowered Enterprise Knowledge Graph with Attention Mechanism

Yadong Shi — 2025-05-27

Enterprise knowledge graphs (EKGs) are pivotal in structuring and analyzing vast amounts of enterprise data, yet conventional construction methods struggle to efficiently capture complex relationships and dynamic enterprise contexts. This paper proposes a Deep Learning (DL)-based enterprise knowledge graph framework that integrates transformer-based architectures, graph attention networks (GATs), and reinforcement learning to enhance the construction, refinement, and querying of EKGs. Specifically, we employ a business-enhanced RoBERTa (BERTO) model for entity and relation extraction from unstructured data, a graph attention network for refining edge weights, and a reinforcement learning agent to adaptively update relationships based on user feedback. Additionally, a query-aware attention mechanism is incorporated for context-sensitive knowledge retrieval. Simulation results demonstrate that the proposed scheme outperforms conventional knowledge graph (GK) and deep learning (DL) models in predictive accuracy, especially under varying signal-to-noise ratio (SNR) conditions. Numerical comparisons reveal that at 10 dB SNR, the proposed scheme achieves a prediction accuracy of 0.74, surpassing the conventional GK (0.49) and conventional DL (0.34) methods. These results underscore the effectiveness of the proposed framework in improving accuracy, adaptability, and scalability in enterprise knowledge management.