HighTech and Innovation Journal
Not a member yet
317 research outputs found
Sort by
Analyzing Online News Dissemination Patterns via Social Network Hypergraph Model
Full text linkThis study aims to develop a novel method for analyzing the complex dissemination patterns of online media news using a social network hypergraph model, addressing the limitations of traditional graph models in capturing many-to-many relationships in news dissemination. The author integrates news content, user nodes, and topic tags into a multi-dimensional hypergraph structure. The approach includes detailed analysis of key elements of news dissemination across four dimensions (subject, content, channel, and effect), construction of the hypergraph model, and design of mechanisms for extracting dissemination paths and evaluating influencing factors. Experiments were conducted on real-world data from multiple social platforms to validate the method's effectiveness. The results demonstrate that the proposed hypergraph model outperforms traditional models (GCN, GAT, and RF) in terms of accuracy, F1 value, and error control. The model effectively reflects the complex structure and dynamic evolution of news dissemination, revealing significant factors such as user activity, topic sensitivity, and structural entropy. This research offers a new perspective on understanding and optimizing online news dissemination by leveraging the hypergraph model's ability to capture multi-dimensional interactions. It provides a more comprehensive and accurate analysis framework, laying a theoretical foundation for constructing efficient information dissemination models
The Impact of Performance Expectations and Perceived Behavioral Control on Employees’ AI Adoption
Full text linkAs AI technologies rapidly permeate industries, the key challenge for enterprises is no longer whether to adopt AI, but how to ensure employees can strategically and efficiently leverage AI tools to improve work performance meaningfully. This issue spans multiple dimensions, from employees’ performance expectancy regarding AI’s tangible value to their mastery of operations and application contexts, and their perceived behavioral control. It also involves whether organizations provide sufficient resources, training, and institutional support, and whether team culture and social influence foster learning and knowledge sharing. This study integrates Social Cognitive Theory and Expectation-Confirmation Theory to elucidate the critical roles of performance expectancy and perceived behavioral control in the AI adoption process and to examine how organizational support and social influence affect AI usage performance through these psychological mechanisms. In addition, we assess the moderating effect of creative self-efficacy on AI adoption. Using survey data from 392 technology-sector employees, we conduct an empirical analysis using structural equation modeling. The results indicate that social influence has a greater impact than organizational support. Performance expectancy is the key mediating variable through which AI use enhances work performance. Moreover, creative self-efficacy amplifies the positive effects of managerial support and social influence on performance expectancy and perceived behavioral control. These findings deepen the theoretical foundation of AI adoption and provide practical guidance for enterprises seeking to improve organizational performance and employee productivity through AI technology
Design of a Cost-Effective Educational Unmanned Ground Vehicle Platform with an Auxiliary Computer
Full text linkThis paper presents the development of a cost-effective, modular, and easy-to-assemble educational unmanned ground vehicle (UGV) system designed for hands-on robotics instruction for high school students. Its methodology incorporates frame redesign using CAD and 3D printing, software integration with DroneKit and Ardupilot, as well as the design of activity-based learning modules. Various performance evaluations, including incline testing, Aruco marker performance tests, and focus testing with students, highlighted successful system operation, system engagement, and learning improvements. The UGV could handle slopes of up to 25 degrees, and vision-guided marker tracking worked with precision. Student feedback was positive, with average Likert scale results of 4.63 for excitement and 4.42 for ease of use. Comparative surveys showed increased user satisfaction with the improved design, though wiring organization, GPS accuracy, and occasional snap-fit difficulties were noted for refinement. A two-tailed t-test showed no change in student interest after testing, but many indicated increased confidence if robotics were further offered in senior high school. The novelty and contribution of this study lie in the integration of a snap-fit 3D-printed modular frame, accessible hardware, autonomous capabilities, and curriculum-oriented learning modules, making robotics education more affordable, engaging, and practical for schools with limited resources
Evaluation and Analysis of Regional Agricultural Eco-Efficiency and Agricultural Economy by the DEA Model
Full text linkObjectives: This paper aims to assess the agricultural ecological and economic efficiency of the Yangtze River Economic Belt by using the data envelopment analysis (DEA) model to evaluate the regional agricultural level. Methods: Relevant data from 11 provinces and cities in the Yangtze River Economic Belt from 2010 to 2020 was collected from statistical yearbooks. Then, the agricultural eco-efficiency and economic efficiency were evaluated using the slack-based measure (SBM) model in the DEA model. Findings: The evaluation result of agricultural eco-efficiency was consistently higher than that of ecological efficiency. From a regional perspective, the eco-efficiency of the downstream area was higher than that of the middle and upper reaches. From the perspective of group division, only Guizhou and Chongqing had a high eco-efficiency. Improvement: The findings suggest that the overall agricultural eco-efficiency in the Yangtze River Economic Belt is low, and there is still a large space for development. It is necessary to further reduce agricultural carbon emissions and non-point source pollution and improve agriculture through technological innovation and other means
Tourist Destination Recommendations Using Deep Learning
Full text linkPersonalized tourist attraction recommendations present a challenging problem in intelligent travel planning. Bangkok, the capital of Thailand, is a popular tourist destination offering a convenient metro system that enables travelers to plan their journeys easily. Leveraging this infrastructure, this study proposes a deep learning-based model designed to classify tourists into five categories: Nature Tourists, Cultural Tourists, Shopping Tourists, Historical Tourists, and Industrial Tourists. The model employs Neural Collaborative Filtering (NCF), utilizing deep neural networks to capture complex, non-linear patterns between users and destinations, surpassing the limitations of traditional matrix factorization methods. It integrates both user-related data, such as tourists’ opinions on destinations, and location-based data from the attractions themselves. To evaluate the model, data were collected from 30 stations along Bangkok's Pink Line, covering the northern part of the city and Nonthaburi province, and 31 tourist attractions along the route. Experimental results demonstrate high classification accuracy across tourism types: 96.26% for Nature Tourists, 80.59% for Cultural Tourists, 93.78% for Historical Tourists, 70.35% for Industrial Tourists, and 97.66% for Shopping Tourists. Furthermore, the study proposes three optimized travel routes tailored to tourist preferences: one for Nature and Cultural Tourists, another for Cultural Tourists, and a third for Historical and Cultural Tourists. By categorizing tourists based on their interests and recommending destinations accordingly, the model supports more informed and personalized travel decision-making. However, this current study serves as a prototype model and can be further applied to problems related to public transportation systems, such as deployment in mobile applications and integration with GPS positioning systems to enhance convenience and accuracy in providing tourist destination recommendations
Hybrid Time Series Methods and Machine Learning for Seismic Analysis and Volcano Eruption Predict
Full text linkVolcanic eruption refers to a natural catastrophe on Earth that poses imminent danger to communities surrounding volcanoes. Therefore, ongoing monitoring of volcanic processes is crucial for effective analysis and observation of volcanic activities preceding an eruption. In response to this, the study presents a novel hybrid time series approach, integrated with machine learning techniques, to enhance the identification and classification of seismic events associated with volcanic eruptions. In this case, time series techniques, including STA/LTA, template matching, and autocorrelation, were implemented to facilitate the detection and classification process. The challenges, however, lie in addressing noise and ensuring accuracy in the analysis of seismic signals. To resolve this, a new hybrid time series method was proposed to improve signal analysis accuracy by integrating multiple time series techniques. In practice, the dataset was collected from Mount Merapi in Indonesia between 2019 and 2021, consisting of a compilation of seismic data categorized by event type, thus enhancing classification accuracy. On top of that, prior to implementing machine learning techniques for signal classification, the hybrid method was employed to efficiently remove noise, ensuring that genuine seismic events were clearly distinguished from spurious signals. Notably, the experimental learning rate was set at 0.01. The results demonstrated that the proposed hybrid method outperformed stand-alone time series techniques, achieving an accuracy of 0.93 to 0.95. This signifies the effectiveness of precise seismic event recognition and categorization, greatly enhancing the volcano monitoring system. Furthermore, the findings offer substantial improvements in the forecasting and risk mitigation associated with volcanic eruptions, hence, advancing reliable seismic analysis methodologies. Ultimately, the method enhances hybrid methods and machine learning for seismic event analysis and volcano monitoring. Doi: 10.28991/HIJ-2025-06-01-08 Full Text: PD
Leveraging Image Analysis and Deep Convolutional Neural Networks for Cutting-Edge Malware Detection and Mitigation
Full text linkIn this study, we investigate using deep learning, i.e., deep convolutional neural networks (DCNNs), for malware detection leveraging network traffic data. Signature-based detection techniques are now proven unable to cope with the extremely high rate of malware variants' evolution. For this reason, this research suggests a novel method of turning raw network traffic data input (APKS, CSVS, and PCAPS) into visual representations for better malware classification. The study trains a model using DCNNs and refines it using the VGG19 architecture and extra convolutional layers to achieve higher detection rates utilizing the CICAndMal2017 dataset. The key metrics of precision (98.5%), recall (99.4%), and F1 score (98.8%) are all observed with a high performance, along with the AUC of 0.93 and accuracy rate of 99.35%. Deep learning is demonstrated to be effective in detecting malware via image-based features, and there is a significant improvement compared to traditional approaches. The novelty in this work is the use of deep learning for malware detection via visual representations of network traffic. Future work will improve computational efficiency, extend the approach to dynamic environments, and learn to be more robust to evasion tactics through adversarial training
Closing Price Prediction of Cryptocurrencies BTC, LTC, and ETH Using a Hybrid ARIMA-LSTM Algorithm
Full text linkThis study aims to develop a hybrid algorithm using the ARIMA model and LSTM-type recurrent neural networks to predict the closing prices of the cryptocurrencies BTC, LTC, and ETH. The methodology includes an exploratory data analysis, followed by the design, implementation, and evaluation of each individual algorithm as well as the combined hybrid algorithm. The results, after experimentation and evaluation of metrics on the test set, indicated that the ARIMA model was inefficient in predicting the closing prices of cryptocurrencies. On the other hand, the hybrid model for BTC showed significant statistical differences in the metrics, with MAE = 729.35 and MAPE = 1.76%. These results indicate better performance from the hybrid model. Regarding the RMSE metric, the hybrid model scored 1157.47, while LSTM scored 1159.99; although statistically equivalent, the hybrid model was numerically better. For the remaining metrics and other cryptocurrencies, both methods were statistically equivalent. For five-day-ahead predictions, the hybrid algorithm continued to yield better results for LTC and ETH
The Rising Cost of Cyberattacks: Trends and Impacts across Industries
Full text linkCybersecurity incidents have escalated sharply since 2020, exposing organizations to mounting financial and operational risks. This study quantifies multi-year trends in five major attack classes, calculates the compound annual growth rate (CAGR) of breaches, and evaluates how targeted security spending mitigates losses across eight industries. Secondary data were extracted from authoritative sources (IBM, ENISA, and Ponemon). Descriptive statistics charted incident growth; Pearson correlation assessed the linkage between phishing volume and breach frequency; ordinary least-squares regression measured the effect of network, infrastructure, and identity-access investments on breach counts. Breaches rose at a 28.3% CAGR from 2020 to 2023. Healthcare incurred the highest mean cost per incident (USD 10.9 million in 2023). Phishing volume strongly correlates with breaches (r = 0.97, p < 0.05), while greater outlays on network and infrastructure security were significantly associated with lower breach rates (β = –0.18 and –0.22, respectively; p < 0.05). Unlike prior sector-specific studies, our cross-industry analysis blends global data with inferential modelling, producing actionable benchmarks that help decision-makers allocate limited cybersecurity budgets where they reduce risk most
Revolutionizing Hospitality: Unraveling the Transformative Potential of Big Data in Tourism and Hotel Management
Full text linkObjective: The purpose of this research is to investigate how big data analysis may be used in the tourist and hotel sectors to improve customer happiness and spur corporate expansion. The goals are to analyze traveler behavior and preferences, derive actionable insights from a variety of data sources, and design customized strategies to enhance customer experiences and promote brand loyalty. Methods/Analysis: To ensure precision and comprehensiveness, the approach incorporates rigorous preprocessing procedures for data. This technique is essential for providing precise insights into the customers, both explicit and implicit. The study provides a thorough understanding of consumer interactions and preferences by including data from social media, travel websites, and hotel booking systems. Findings: The research offers significant insights that demonstrate the capacity to improve consumer experiences, tailored products, optimized services, and effective marketing tactics. The results emphasize how important it is to understand client preferences to inform corporate strategy and create a competitive edge. Conclusion: The potential of big data analysis in the travel and hospitality sectors is shown in this research, which adds to the rapidly developing subject. This study highlights how big data analysis plays a critical role in enhancing the tourist experience and promoting industry innovation by clarifying the relationship between technology and customized services. Doi: 10.28991/HIJ-2025-06-01-014 Full Text: PD