HighTech and Innovation Journal
Not a member yet
317 research outputs found
Sort by
Generative AI for Enhancing Accessibility and Inclusion in Higher Education: A Systematic Review
Full text linkThis study reviews existing literature on generative artificial intelligence (AI) and its accessibility for students with visual, hearing, and motor disabilities in higher education. The objective is to identify gaps in the implementation of inclusive education practices. The PRISMA protocol guided the review process, and the Scopus and Web of Science databases were selected for their recognized academic rigor and comprehensive coverage. The first phase involved the review of 54 articles in English from 2023 to 2024. The selection process involved prioritizing articles based on empirical scientific studies on AI applications for students with disabilities, and discarding articles that did not meet the criteria. Ultimately, only five articles were selected. The findings reveal a significant research gap regarding the role of generative AI in supporting these students. Notably, the selected articles tend to focus more on sensory disabilities than on motor disabilities. This study is pioneering in pointing out the lack of research on motor disabilities during the analyzed period, a key aspect of AI in higher education. These findings underscore the necessity of further research that aligns with the UN 2030 Agenda, specifically Goals 4 (Quality Education) and 10 (Reduced Inequalities), promoting the development of AI tools that foster equal opportunities and inclusive education
Innovative Compositions of Shotcrete Mixtures for Reinforcement of Underground Mine Excavations
Full text linkThe objective of this article is to develop a polymer-modified shotcrete composition to improve underground mine support. The authors propose a new formulation by integrating an aqueous emulsion of SKS-65 GP grade B latex into cementitious matrices. Methods include X-ray diffraction, particle-size analysis, rheological testing, mechanical strength tests, numerical modeling, in-situ trials at the Zholbarysty mine, and statistical evaluation. Findings show a 45% increase in compressive strength and a 30% reduction in rebound loss compared to standard mixtures. Field core samples confirmed reproducibility, with strength values within 1% of those from laboratory-tested cubes. The improved mix allows a 50% reduction in lining thickness, expanding the tunnel cross-section by 5% and lowering operational costs by 39%. Cost-benefit analysis and cross-sectional evaluation validate the approach's efficiency. The novelty of this work lies in combining microstructural insights with field-scale application, clarifying polymer-film formation mechanisms, and presenting an optimized, scalable shotcrete mix design. This integrated method provides a practical and cost-effective reinforcement solution, advancing current shotcrete technologies for underground operations
Multi-Dimensional Analysis of Subjective and Objective Empowerment Methods in Online Civic Education
Full text linkTo enhance the scientificity and effectiveness of online ideological and political education (Cyber Civics), this article aims to construct a multi-dimensional and quantifiable evaluation model. Methodologically, the article starts from the four dimensions of education subject, object, content, and medium, combines subjective empowerment (hierarchical analysis method AHP) and objective empowerment (entropy power method), and introduces an intelligent optimization algorithm - the long-nosed Cuckoo Optimization Algorithm (COA) to optimize the combination of weights, and constructs the COA-Mixed Cyber Ideology and Political Education Evaluation model. The results show that the model is better than the traditional model in terms of weight distribution, with the four-dimensional index weights of 0.358 for the educational subject, 0.245 for the educational object, 0.207 for the educational content, 0.189 for the educational medium, and the maximum composite score of the sample is 0.875, and the optimization coefficient of the model prediction error is α=0.35, which is significantly better than that of GWO-Mixed (α=0.33) and KOA-Mixed (α=0.33). Mixed (α=0.36). It is concluded that multi-dimensional analysis combined with subjective and objective empowerment and intelligent algorithm optimization can more objectively and accurately assess the effectiveness of online ideological and political education, which provides a feasible path and theoretical support for improving the quality of ideological and political education in colleges and universities
Predicting Adolescent Suicide Risk in Smart Cities: An AI-Driven, Privacy-Preserving Architecture
Full text linkThis study aims to improve the accuracy, speed, and safety of suicide risk assessment among adolescents in the digital ecosystems of smart cities. To achieve this goal, an integrated system architecture was developed that combines natural language processing methods, transformer models, and privacy-preserving computation. The methodological part includes large-scale textual data analysis, distributed processing in Apache Spark and Hadoop environments, and the use of federated learning, which allows models to be trained without transferring sensitive source information. The evaluation was conducted on open mental health datasets and supplemented by a series of experiments simulating the system's operation in real time, as well as surveys of specialists – psychologists, educators, and IT experts. The analysis showed that transformer models, particularly BERT, significantly outperform classical algorithms, achieving an AUC-ROC of 0.96 and an F1 score of 0.92 with an average response time of 2.4 seconds. Survey participants noted the importance of transparency and data protection, and the proposed architecture received high marks for reducing the risk of information leaks and providing robust audit mechanisms. The novelty of the work lies in the combination of predictive analytics, federated learning, differential privacy, and blockchain traceability in a single application-oriented system. The results show that ethically sound and rapid suicide risk detection can be implemented in schools, medical institutions, and municipal services, providing both practical benefits and contributing to methodological advancements
Analyzing Urban and Rural Water Pollution Impacts with an Integrated Ecological Governance Model Approach
Full text linkFrom the perspective of aquatic ecology, there are problems of insufficient analysis and poor governance effect on the impact and ecological governance of urban and rural water pollution. Therefore, to achieve a good water cycle, it is of great practical significance to analyze the impact of urban and rural water pollution from the perspective of aquatic ecology and study the ecological governance model. Taking a certain research area as an example, an integrated ecological governance model under the perspective of aquatic ecology is designed, including source control, pollution interception, and restoration. The impact of urban and rural water pollution on soil environment, groundwater environment, and agricultural environment is analyzed, and the change of water pollution concentration before and after application is studied. The research results show that in the soil environment of the research area, most areas are lightly polluted, two other areas (Area 3, Area 4) are heavily polluted, one area (Area 7) is moderately polluted, and one area (Area 11) is unpolluted; in terms of groundwater environment, the degree of groundwater pollution in Area 1 and Area 2 is the highest, followed by Area 3 and Area 4, then Area 7 is moderately polluted, and other areas are lightly polluted or unpolluted; in terms of agricultural environment, as the pollution degree of irrigation water source increases, the emergence rate, yield, and dry matter content of crops all show a decreasing trend, indicating that the more serious the water pollution, the more serious the impact on the agricultural environment; after applying the research method, the highest water pollution concentration has been reduced by 0.8 mg/L, and the overall data is below 1.0 mg/L, the water pollution concentration has been reduced. Through this research, it is expected to achieve in-depth ecological governance and protect the aquatic environment. Doi: 10.28991/HIJ-2025-06-01-04 Full Text: PD
Techno-Economic Evaluation of Carbon Capture and Storage for Combined Cycle Power Generation
Full text linkCarbon dioxide (CO₂) is a major driver of greenhouse gas emissions, which lead to an increase in Earth's temperature and subsequently drive climate change. CO₂ is primarily produced from fossil fuel-based power generation. Carbon capture and storage (CCS) is a CO₂ capture technology that can be added to fossil fuel power generation. This study evaluates the he technological, financial, and ecological impacts of upgrading CCS technology on a Natural Gas Combined Cycle (NGCC) power generation with three blocks. Amine-based post-combustion capture technology is applied in this study. Simulations were performed employing the Integrated Environment Control Model software. The addition of CCS significantly reduces net power output across all blocks. For Block 1, net power declines from 133 MW to 97.6 MW, a 27% reduction, while Block 2 drops by 17%, from 441.7 MW to 368.1 MW. Block 3 shows a 13% decrease, with net power falling from 441.9 MW to 385.5 MW. Thermal efficiency also declines with the installation of CCS. Corresponding efficiency losses are also notable: Block 1 falls from 40.85% to 30%, Block 2 from 45.24% to 37.69%, and Block 3 from 53.89% to 46.79%. The levelized cost of electricity increases considerably alongside CCS implementation, rising by 80% for Block 1 (0.0843 to 0.1522 USD/kWh), 47% for Block 2 (0.0761 to 0.1114 USD/kWh), and 42% for Block 3 (0.06618 to 0.0874 USD/kWh). Sensitivity analysis indicates that LCOE competitiveness with the national weighted average is achievable when carbon prices exceed 145 USD/t CO₂ for Block 1, 90 USD/t CO₂ for Block 2, and 45 USD/t CO₂ for Block 3. These findings emphasize the trade-offs between power generation efficiency, costs, and carbon capture, providing essential insights for future energy policy and CCS adoption strategies
A Review on Federated Learning on Sensor-Based Human Activity Recognition
Full text linkDeep learning has demonstrated exceptional human activity recognition (HAR) performance by extracting complex features from inertial data. However, this centralized training approach aggregates data from multiple user devices into a central server and raises significant privacy concerns. Federated learning (FL) is proposed as an alternative. It provides a privacy-preserving scheme by training data analytics models on local users’ devices rather than transferring raw data to a central server for data processing. Although FL is widely applied to various pattern recognition applications, its use in sensor-based HAR is limited, and reviews of the HAR application are even scarcer. Therefore, this paper provides a comprehensive review of FL in HAR. This paper analyzes FL’s architectural design, data model training strategies, and model aggregation techniques. A comparative analysis between FL-based and machine learning methods is presented. The challenges, including data heterogeneity, data privacy, and communication costs, are identified through the findings, while the potential research direction of FL in HAR is underscored. This paper provides insights into the current state of FL for HAR, pinpoints research gaps, and outlines encountered challenges and potential research directions
Research on Carbon Emission Estimation of Rural Tourist Attractions Through Digital Management
Full text linkObjectives: This study aims to estimate the carbon emissions of scenic spots in rural tourism using digital management technology. Methods: The Dashahe National Wetland Park, located along the old course of the Yellow River in Feng County, Jiangsu Province, was taken as a case for analysis. During the analysis process, the carbon emission, carbon absorption, and net carbon emission amount of the park during 2018-2023 were estimated. The correlation between different types of land area and the carbon absorption amount was analyzed. Findings: The carbon emission of the wetland park increased annually, but the carbon absorption amount also showed a consistent upward trend, resulting in relatively stable net carbon emissions over the study period. Moreover, the area of wetlands, water bodies, and grasslands exhibited a significant positive correlation with the carbon absorption amount, whereas the correlation between the area of cultivated lands and garden lands and the carbon absorption amount was insignificant. Innovation: This research applied digital management technology to precisely collect data related to carbon emissions within the scenic spot, enabling a more reliable estimation of its carbon footprint
FATA-ResNet Network for CAD/CAM Integration in Cloud Manufacturing
Full text linkThis paper focuses on the application of mechanical engineering CAD/CAM integration technology under the cloud manufacturing framework, aiming at solving the current technical integration problems in manufacturing informatization. The study analyzes the demand and current situation of 3D CAD/CAM integration in a cloud manufacturing environment, combines the mirage optimization algorithm (FATA) and residual neural network (ResNet), and proposes a CAD/CAM integration application analysis model based on the FATA-ResNet network. Firstly, the functional requirements of CAD/CAM technology integration in a cloud manufacturing platform are clarified, including 3D model uploading and downloading, process file generation, and cross-platform data sharing. Then, the hyperparameters of the ResNet network are optimized by the FATA algorithm to improve the accuracy and efficiency of the model in integration application analysis. The experimental results show that the FATA-ResNet model outperforms the traditional model in terms of accuracy, recall, and F1 score while possessing faster convergence speed and higher computational efficiency. In addition, the operation modules in the cloud platform, including the task management interface and 3D process editing function, were designed and validated, further demonstrating the practicality of the method. Future research will focus on the validation of multi-scene data, model resource optimization, and real-time collaborative operation to promote the in-depth application of CAD/CAM technology in intelligent manufacturing and provide support for the digital and intelligent development of manufacturing
A Data-Driven Adaptive Scheduling Framework for Vehicle Maintenance Using Deep Reinforcement Learning
Full text linkThis paper proposes a data-driven adaptive scheduling method based on the Deep Deterministic Policy Gradient (DDPG) algorithm to address the challenges that traditional vehicle dynamic maintenance scheduling methods struggle to cope with real-time, complex and resource optimization issues. A mathematical model of vehicle dynamic maintenance scheduling is constructed, defining the state space, action space and reward function. Then, the DDPG reinforcement learning framework is used to optimize strategies through the Actor-Critic structure. Contrastive experiments are also carried out in a simulation environment to evaluate the algorithm's performance. The results indicate that the DDPG algorithm achieves an average maintenance response time of 23.4 minutes, approximately 34% shorter than the genetic algorithm. Its resource utilization reaches 88.7%, over 13% higher than traditional methods. Moreover, the maintenance satisfaction score is 4.6 out of 5. The findings show that the algorithm has remarkable advantages in multi-objective scheduling optimization and provides feasible paths and technical support for the intelligence of vehicle dynamic maintenance systems