Data and Metadata (Journal)
Not a member yet
727 research outputs found
Sort by
Hybrid Intrusion detection model-based density clustering approach and deep learning for detection of malicious traffic over network
Full text linkIntrusion detection in modern network environments poses significant challenges due to the increasing volume and complexity of cyber-attacks. This study proposes a hybrid approach integrating density-based clustering with deep learning to identify malicious traffic over the network. The proposed framework consists of two steps: clustering and classifying data. in clustering, the proposed model uses density clustering techniques to pre-process and segment network traffic into coherent clusters, thereby reducing data noise within clusters. The deep learning model analyses these clusters, accurately distinguishing between benign and malicious activities. The proposed model was tested over the benchmark dataset CIRA-CIC-DoHBrw-2020. The performance of the proposed model compared with standard machine learning models and the number of states of the artworks. The experiment result demonstrates that our hybrid model significantly improves detection accuracy and reduces false-positive rates compared to existing methods
Effect of 3D printing parameters on the mechanical characteristics of carbon fiber- reinforced PLA
Full text linkThe comparative results of the mechanical behavior of carbon fiber-reinforced Polylactic Acid (PLA FC) specimens of two brands of filaments for printing by the Fused Deposition Modeling (FDM) process are presented. The experiments were carried out according to ASTM D638 14, using Type I specimens with the established dimensions. For the generation of the 3D model, parameters such as printing temperature, printing speed, density, and filling pattern were set. Cubic, gyroid, and triangular filling patterns were used, with filling densities of 40%, 60%, and 80%. For each configuration, a G-code was generated and used for the fabrication of each specimen. A total of 90 specimens were used, which were divided into two groups according to the brand. Subsequently, tensile tests were carried out to determine the mechanical properties by analyzing the stress-strain curves under the established conditions. Comparative analysis revealed that SUNLU\u27s PLA FC filament achieves higher ultimate stress values, while Artillery\u27s filament has a better ability to withstand deformation. Likewise, the filler pattern that withstood the greatest load was the cubic one.
Microclimate condition monitoring system for the prevention of methane contamination in the methane contamination in compost production in Microfarms
Full text linkThis work focuses on the development and implementation of a microclimate variable control system to prevent microbial contaminants in compost production, with the objective of investigating composting methods and how they can help reduce the production of methane, a greenhouse gas, and thus contribute to environmental care. The Action-Research methodology is used with the use of sensors that monitor data on environmental variables of ambient temperature, relative humidity and soil moisture, which are sent to an IoT platform where the necessary data are processed and generated. A specific infrastructure is designed for compost production, which includes a closed box lined with greenhouse plastic, a container for the compost, a piping system to maintain humidity, a heater to raise the temperature and a protective box for the sensors. Also included is the development and training of a neural network model that predicts methane production based on the above variables. The data show that composting at temperatures between 55-65 degrees Celsius, using aerobic biological methods, significantly reduces methane production by eliminating bacteria responsible for methane generation. The data collected and model predictions can be monitored remotely through the IoT platform. At the conclusion of the work, the compost generated was found to be suitable for micronization
Comparison of tools for the creation of VLO for the subject of Algorithms. UTN case
Full text linkThis paper compares the performance of three authoring tools, eXeLearning, H5P and Xerte, used to create Virtual Learning Objects (VLOs). The research stems from the need to identify the most appropriate tool for creating VLOs. The study aimed to evaluate these tools through an objective comparison based on 15 pre-defined criteria. For this purpose, a methodological approach of the UP4VED methodology for virtual environments was used in addition to implementing the ADDIE methodology for developing VLOs. The results showed that eXeLearning achieved the highest score with 10.38 points, followed by Xerte with 8.83 and H5P with 7.74. Subsequently, DeLone & McLean\u27s success model was applied to assess students\u27 perceptions of using OVAs in an online course, with a minimum favourability index of 81.92% for quality of service and a maximum of 92.80% for quality of information. These results confirm the acceptability and effectiveness of OVAs as digital resources that enhance self-learning in educational environments
Development and validation of a new artificial intelligence tool (GeneClin) for the clinical diagnosis of genetic diseases
Full text linkIntroduction: Advances in the field of Artificial Intelligence (AI) and Machine Learning (ML) have considerable potential to improve the diagnosis and management of rare genetic diseases, due to the human inability to memorize information on a multitude of these diseases, which AI tools could store, analyze and integrate. Objective: to develop and validate a new AI tool for the clinical diagnosis of genetic diseases. Methods: A prospective, cross-sectional, analytical, observational study was conducted at the application level, with a qualitative-quantitative approach and contributing to a technological development project. It was characterized by four stages: selection of the AI tool, selection of the knowledge base, development of the virtual assistant, validation process and implementation in the clinic. Results: A total of 246 patients with genetic diseases and congenital defects were evaluated. The most predominant genetic category was monogenic genetic syndromes with 223 patients who attended the consultation (90.7%). A success rate of 84.1% was obtained and a success/no success ratio of 4.34. The highest percentage of successes was achieved in monogenic or Mendelian syndromes. There were no significant differences between successes and failures in both chromosomal aberrations and congenital defects of environmental etiology. Conclusions: Through this research, an AI virtual assistant has been validated for the clinical diagnosis of genetic diseases with a high percentage of effectiveness of 84%, which confirms its usefulness to support the clinical diagnosis of cases with genetic diseases
Computer Vision for Vehicle Detection: A Comprehensive Review
Full text linkThe rapid increase in vehicle numbers has exacerbated challenges in modern transportation, leading to traffic congestion, accidents, and operational inefficiencies. Intelligent Transportation Systems (ITS) leverage computer vision techniques for vehicle detection, improving safety and efficiency. This paper aims to provide a comprehensive review of vehicle detection methods in ITS. Traditional image-processing techniques, including Scale-Invariant Feature Transform (SIFT), Viola-Jones (VJ), and Histogram of Oriented Gradients (HOG), are analyzed. Additionally, modern deep learning-based approaches are examined, distinguishing between two-stage methods such as R-CNN and Fast R-CNN, and one-stage methods like YOLO and SSD. Various image acquisition techniques, including Mono-vision, Stereo-vision, Thermal/Infrared Cameras, and Bird’s Eye View, are also reviewed. The analysis highlights the evolution from handcrafted feature-based methods to deep learning techniques, demonstrating significant improvements in detection accuracy and efficiency. One-stage detectors, particularly YOLO and SSD, offer real-time performance, while two-stage methods provide higher precision. The impact of different imaging modalities on detection reliability is also discussed. Advances in deep learning and imaging techniques have significantly enhanced vehicle detection capabilities in ITS. Future research should focus on improving robustness in diverse environmental conditions and optimizing computational efficiency for real-time deployment
Revolutionizing Blood Banks: AI-Driven Fingerprint-Blood Group Correlation for Enhanced Safety
Full text linkIdentification of a person is central in forensic science, security, and healthcare. Methods such as iris scanning and genomic profiling are more accurate but expensive, time-consuming, and more difficult to implement. This study focuses on the relationship between the fingerprint patterns and the ABO blood group as a biometric identification tool. A total of 200 subjects were included in the study, and fingerprint types (loops, whorls, and arches) and blood groups were compared. Associations were evaluated with statistical tests, including chi-square and Pearson correlation.The study found that the loops were the most common fingerprint pattern and the O+ blood group was the most prevalent. Discussion: Even though there was some associative pattern, there was no statistically significant difference in the fingerprint patterns of different blood groups. Overall, the results indicate that blood group data do not significantly improve personal identification when used in conjunction with fingerprinting.Although the study shows weak correlation, it may emphasize the efforts of multi-modal based biometric systems in enhancing the current biometric systems. Future studies may focus on larger and more diverse samples, and possibly machine learning and additional biometrics to improve identification methods. This study addresses an element of the ever-changing nature of the fields of forensic science and biometric identification, highlighting the importance of resilient analytical methods for personal identification
Improving Student Graduation Timeliness Prediction Using SMOTE and Ensemble Learning with Stacking and GridSearchCV Optimization
Full text linkIntroduction: Timely graduation is a key performance indicator in higher education. This study aims to improve the accuracy of predicting student graduation timeliness using ensemble machine learning techniques combined with SMOTE and hyperparameter optimization.Methods: This is a quantitative predictive study. The population includes students and alumni of Universitas Islam Riau. A sample of 160 respondents was obtained via purposive sampling. Data were collected using structured questionnaires encompassing academic variables (e.g., GPA, credits, attendance) and non-academic variables (e.g., stress, social support, extracurricular activity). After preprocessing and label encoding, SMOTE was applied to balance class distribution. Several classifiers (Naïve Bayes, SVM, Decision Tree, KNN) were tested, with ensemble learning (voting and stacking) implemented and optimized using GridSearchCV.Results: The stacking ensemble model optimized with GridSearchCV achieved the highest performance with an accuracy of 99.37%, precision and recall above 0.99, and minimal misclassification. This outperformed individual models and previous approaches in the literature. Conclusions: The integration of SMOTE, ensemble methods, and GridSearchCV significantly enhances predictive accuracy for student graduation timeliness. The resulting model provides a robust framework for academic risk detection and early intervention
Customer Sentiment Analysis for Food and Beverage Development in Restaurants using AI in Jordan
Full text linkIntroduction: customer sentiment analysis is a vital tool for understanding consumer preferences and enhancing service quality in the food and beverage industry. Online reviews significantly influence customer decisions, making it essential for businesses to analyze sentiment trends and manage their digital reputation effectively. This study examines customer sentiment across different establishment types and digital platforms in Jordan, providing insights into sentiment patterns and their strategic implications.Method: a dataset of 384 customer reviews from various restaurants and hotels was analyzed using a rule-based sentiment classification approach. Sentiments were categorized as positive, neutral, or negative. To assess sentiment variations, an ANOVA test was conducted to compare establishment types, and a Chi-Square test was performed to examine differences across digital platforms.Results: findings indicate that luxury hotels and fine dining establishments receive more positive sentiment, while budget hotels and fast food chains experience higher negative sentiment. However, the ANOVA test showed no statistically significant sentiment differences across establishment types, suggesting that all businesses receive a mix of sentiment categories. The Chi-Square test confirmed significant sentiment differences across platforms, with TripAdvisor attracting the most positive reviews, Facebook and Google Reviews showing balanced sentiment, and Twitter experiencing the highest negative sentiment.Conclusion: these findings emphasize the importance of platform-specific digital reputation management. Businesses should strategically engage with customers on different platforms, address complaints proactively, and utilize AI-driven sentiment analysis tools to improve customer satisfaction. Future research should explore AI-based predictive analytics and sentiment monitoring for enhancing service quality in the hospitality industry
Methodological Proposal for the Design and Validation of Research Instruments Supported by Artificial Intelligence
Full text linkIntroduction: the validity of data collection instruments is essential to ensure the quality and replicability of scientific studies; traditional methods require time, resources, and expert participation, making validation difficult. Objective: To develop a procedure for the design and validation of research instruments using Artificial Intelligence as a methodological support tool. Methods: an eight-phase model was designed, ranging from conceptual review and item formulation to linguistic evaluation, simulated rational validation, comprehension verification, internal consistency analysis, and structural optimization.Results: the process demonstrated applicability, technical coherence, and practical utility. ChatGPT 4.5 enabled the automation of analyses and the generation of content aligned with theoretical constructs, optimizing the preliminary validation phases. Conclusions: AI represents a viable alternative in resource-limited settings. While it does not replace classic empirical methods, it complements methodological rigor in key stages. Ethical and technical protocols must be established for its responsible use in scientific research