International Journal of Applied Science

Generative AI (GAI) Use for Cybersecurity Resilience: A Scoping Literature Review

Jessica Parker — 2025-03-09

With cyberattacks increasing in volume and number, organizations are increasingly at risk of adverse financial and reputational impacts. Cyber attackers are quick to implement technologies like Generative Artificial Intelligence (GAI) to enhance attacks, while organizations have yet to fully benefit from GAI to improve cybersecurity defenses. This scoping literature review analyzes current research and identifies gaps in the literature about how Generative Artificial Intelligence (GAI) can be used to enhance cybersecurity resilience. The analysis includes an overview of GAI, ethical considerations and challenges, future directions and research opportunities, and a discussion of how this GAI research can be applied.

Application of Machine Vision in Structural Deformation and Health Monitoring

Zhonglai Qin — 2025-04-06

There are two main methods of structural health monitoring. Among them, the traditional structure health monitoring technology is mainly artificial, which has some problems such as inflexibility, large error, high cost and difficult to adapt to the change of natural environment. The structural health monitoring method based on machine vision has the characteristics of flexible measuring point, high accuracy, fast speed and no contact. A variety of machine vision technologies, such as image acquisition, image processing, three-dimensional vision and deep learning technologies, have made great progress, and their application scenarios are constantly expanding. This paper is based on infrared image acquisition technology, 3D vision acquisition technology, image stitching technology, stereo vision technology four perspectives. The improvement of machine vision technology in the field of structural health monitoring is described in detail. The development trend of machine vision acquisition technology and image processing technology and the huge growth space of machine vision technology are analyzed. The application range of these key technologies is also introduced. The application results in building crack detection, seepage detection and fire detection are summarized, and the future development of this technology is prospected from the aspects of algorithm robustness.

An Enhanced Framework for Urban Water Consumption Analysis: Feature Clustering with Ensemble Methods

Faye F.F. Jiang — 2025-04-06

Urban water consumption analysis presents significant challenges due to the complex interplay of socioeconomic, demographic, and built environment factors. This paper introduces a novel Feature Clustering Framework of TopK and Threshold with Ensemble Method (FCTTE) specifically designed to address high-dimensional urban datasets. We evaluate this framework using a comprehensive dataset of 1,120 features across eight domains related to New York City's urban environment. Our experiments demonstrate that FCTTE significantly outperforms conventional feature selection methods, improving LightGBM classification accuracy by 4.6% compared to baseline, while traditional methods achieved only 1% improvement. The framework identified median family income, energy usage intensity, adult male population, greenhouse gas emissions, and commercial building characteristics as the most influential factors affecting water consumption. By effectively managing feature redundancy through hierarchical clustering and strategic selection, FCTTE provides urban planners with interpretable insights for water resource management while maintaining superior predictive performance. This integrated approach bridges the gap between fragmented analyses of individual urban factors and the need for holistic understanding of water consumption patterns in complex urban environments.

Pyrolysis Reaction and Kinetic Analysis of Model Compounds

Yangchen Sun — 2025-04-16

In this paper, Pearson correlation analysis was used to correlate the different mixing ratios of the three pyrolysis combinations with the associated pyrolysis product yields. Then, a paired-sample t-test model was used to explore whether the catalyst DFA plays an important catalytic role in promoting the pyrolysis of CS, CE and LG. In order to address the effect of mixing ratios of pyrolysis combinations on the yields, this paper plotted line graphs with the mixing ratios of pyrolysis combinations as the horizontal axis and the pyrolysis gas yields as the vertical axis, and interpreted the effect of mixing ratios on the yields of each gas according to the trend of each gas.

A Design of Bionic Gripper to Enhance the Persistence of Wild Life Monitoring Performed by Drone

Yuheng Zhang — 2025-04-18

In recent day, drones has become an essential tools in environmental observation and wildlife animal monitoring. It is undeniable that the effectiveness of this tool is often limited by challenges in landing and energy efficiency, particularly in dense forest environments. To address this potential problems, this study proposes a bionic gripper which is inspired by bird talons for enhancing drone landing capabilities, enabling a stable landing on complex environments for example tree branches. The gripper features a self-locking mechanism, which highly reduces the energy consumption while maintaining a stable and reliable holding. Using SolidWorks and Fusion 360, the gripper was designed and stimulated for a lightweight and adaptive performance. A key application of this technology is in the conservation for endangered species such as crested-ibis. This system is allowing drones to conduct long-term, low-disturbance observation, which not only extends the operation time but also minimizes the interference from noises, that contributes to a more sustainable and effective wildlife research as well as environmental monitoring.

Optimisation of Spraying Parameters for Boom Sprayers

Dong Liang — 2025-04-21

In order to investigate the influence of factors on the spray deposition of the spray bar sprayer, using the designed mobile spray bar spray device, the spray height, spray bar travelling speed and spray pressure were selected as the test factors, and the deposition amount of seductive red per unit blade area Y₁ and the deposition coefficient of variation Y₂ were taken as the evaluation indexes for the test, and the 3-factor, 3-level orthogonal rotary spray deposition test was carried out. The Box-Behnken response surface method was used to analyse and obtain the mathematical model of the spray deposition characteristics of spray height H, spray rod travel speed v and spray pressure P. The influence of each factor on the spray deposition characteristics and the optimal working parameters of the spray machine were obtained according to the regression analysis results. The results show that the forward speed, working height and spray pressure have a significant effect on the spray deposition characteristics, and the degree of influence in the order of the factors from the largest to the smallest is the spray pressure, the working height, the forward speed, the spray pressure, the working height and the spray pressure. The mean value of droplet deposition per unit area is negatively correlated with forward speed and nozzle working height, and positively correlated with spray pressure. The coefficient of variation of deposition is negatively correlated with the spray pressure, positively correlated with the forward speed, and decreases and then increases with the increase of the working height, which indicates that in the case of a certain working height, increasing the spray pressure or reducing the forward speed can effectively improve the uniformity of droplet deposition of the sprayer.

Visible Body Anxiety: The Mediatized Management Practices of Quantifying Body Shapes and Physicality

Xiaotong Shi — 2025-04-26

The proliferation of fitness and exercise-related social media, applications, and wearable devices has provided individuals with convenient tools for body management. These technological means have enabled the mediatized management of bodies, permeating various aspects of daily life. However, this process has also raised tensions regarding the relationship between individuals and media control. This study adopts media visibility and digital rationality as theoretical perspectives, employing qualitative interviews and participatory observation to explore the impact of fitness applications and devices on users' body management practices. It investigates how users navigate the visible/invisible dichotomy of "highlighting" and "concealing" within the data relationships of "quantifying body shapes." The findings reveal that when users face motivational deficits, they often utilize visual body tracking tools to monitor and record bodily data, enabling real-time observation of bodily changes. While "self-quantification" facilitates body management, it also generates new challenges and anxieties. The prevalence of flow-dominated quantification aesthetics leads users to excessively focus on body image and overly rely on technological tools, thereby neglecting the natural state of the body and self-perception, which may adversely affect users' physical and mental health.

Adaptive Containment Control for Nonlinear Multi-Agent Systems with Input Quantization and Sensor Faults

Wen Wen — 2025-04-29

This study investigates the adaptive containment control problem for a class of nonlinear multi-agent systems with input quantization and sensor faults. A state observer and a radial basis function neural network are respectively employed to estimate unmeasurable states and approximate unknown nonlinear functions. An absolute cubic Lyapunov function is designed to compensate for the influence of sensor faults on the systems. A filter is introduced to reduce computational complexity. Adaptive laws are developed to update the estimates of uncertain dynamic parameters, fault coefficients, and the filter-error compensation term. A distributed adaptive control scheme is proposed to ensure that all followers converge to the convex hull formed by the leaders. The stability of the closed-loop system is strictly proved based on stability theory, and the effectiveness of the proposed control method is verified by numerical simulation.

Reservoir Characteristics and Influencing Factors of the Jurassic Yan'an Formation in the Hao Tan Area of the Ordos Basin

Jiahao Wang — 2025-05-22

To study the reservoir characteristics and main controlling factors of Jurassic Yan'an formation in Hao Tan area of Ordos Basin. The reservoir characteristics and its main controlling factors were revealed by the experimental observations such as rock sheet analysis, high pressure mercury pressure experiment and cathodoluminescence. The Jurassic reservoirs in Hao Tan area are mainly located in Yan 10~Yan 8 section.The results show that the main types are feldspathic sandstone and feldspathic quartz sand with medium to low maturity, and the main types of pores include intergranular pores, intergranular dissolution pores and feldspathic dissolution pores. The physical properties of the reservoir are dominated by mesopore mesoturbidity, with medium throat radius and low drainage pressure. The reservoir has experienced moderate to strong mechanical compaction with various types of cementation dominated by carbonate, siliceous and clay minerals, among which the pore structure of the reservoir has been significantly improved by dissolution. The reservoirs of the Yan'an Formation have certain storage capacity and development potential, especially the Yan 10 and Yan 8 reservoirs have better physical properties, and the sedimentation and diagenesis have affected the inhomogeneity of the reservoirs. It is concluded that the Yan 8 reservoir (72% class II) and the local sweet spot of Yan 10 (class I porosity >16%) have development potential, but different development strategies should be formulated for the high permeability zone of Yan 9. The results of the study can provide a theoretical basis for the exploration and development of the same type of Jurassic tight sandstone reservoirs in the Ordos Basin.

Survey on Privacy Preserving in Crowd Sensing

Tian Maoze — 2025-05-30

As an emerging sensing technology, crowd sensing has gained wide attention in many application fields and is developing rapidly. However, with the popularization of crowd participation in sensing tasks, the risk of user privacy leakage is also increasing, which becomes an important problem to be solved urgently. When users participate in sensing tasks, they need to upload personal information or sensor data, which often contains sensitive information. Without effective privacy protection measures, user privacy may be leaked or abused. The core goal of privacy protection is to ensure that users' private information will not be leaked when they participate in the task. This paper analyzes the related research progress of privacy protection in the field of crowd sensing, and summarizes the main challenges currently faced.

Analysis and Prospect of Federated Learning and Privacy Protection Technology

Peng Hongye — 2025-05-30

As a new type of distributed machine learning technology, federated learning has shown great application potential in the Internet of things, health care, smart home, finance and other fields. Its core advantage is that it can conduct model training without centralized data, effectively reducing the cost of data transmission and storage, and avoiding the risk of privacy disclosure. However, with the wide application of Federated learning, the problems of data security and privacy protection are increasingly apparent, especially in the face of complex network attacks and data leakage risks. This paper deeply analyzes the basic principles and architecture of Federated learning, and discusses the possible privacy threats in data transmission, model updating and participating devices in detail. Combined with the existing security protection technologies, such as differential privacy, encryption algorithm and secure multi-party computing, this paper discusses how to effectively ensure the security of Federated learning. Finally, the article also looks forward to the future development trend of Federated learning in privacy protection, model optimization, computational efficiency and cross domain collaboration, aiming to provide theoretical support and practical guidance for the further development and application of this technology.

Generative Graph based Model Inversion Attack on Graph Neural Network

Hongfa Ding — 2025-05-30

Aiming at the privacy leakage risks of Graph Neural Networks (GNNs) in black-box scenarios, this paper proposes a Generation-Graph based Model Inversion Attack on GNN (GenG-MIA). By constructing a generative attack framework and integrating public knowledge distillation with structural optimization strategies, the proposed method effectively addresses challenges such as the high-dimensional sparsity of graph structure data, generative bias, and model collapse. GenG-MIA operates in two stages: first, during the public knowledge distillation stage, Wasserstein GAN is employed to train generators and discriminators on public datasets, enhancing the authenticity and diversity of generated graphs through a diversity loss term and introducing local/global discriminators to mitigate semantic gaps; second, in the structure revelation stage, potential vector projections are optimized to align with the feature space of the target model, thus recovering missing sensitive structures in training graphs. Experimental results show that GenG-MIA significantly outperforms existing methods in terms of attack accuracy and efficiency, enabling the efficient reconstruction of the topological structures of target training graphs and providing a new paradigm for privacy risk assessment of GNN models. This study further expands the application potential of generative attacks in complex graph data scenarios and offers theoretical references for privacy protection and model robustness design.