Sistema Regional de Información
en línea para Revistas Científicas de América Latina,
el Caribe, España y Portugal

The rapid growth of the Internet of Things (IoT) in industrial environments has increased efficiency but also heightened vulnerability to sophisticated cyber-attacks. Traditional cyber security approaches are insufficient to protect critical infrastructure, creating a need for dynamic, adaptive solutions. Evolutionary algorithms (EAs), owing to their ability to explore large search spaces and optimise parameters, offer a promising route to enhancing IoT security. This review highlights the integration of EAs with deep-learning techniques to improve intrusion detection and system resilience. Building on this background, we propose an adaptive cyber-security framework that leverages evolutionary optimisation and continual learning to detect, prevent and mitigate attacks in real time. The study emphasises the importance of validating hybrid models in real-world settings and of optimising computational efficiency. Future work should investigate autonomous response mechanisms and the scalability of solutions for large-scale Industrial IoT (IIoT) deployments, ensuring robust protection against emerging threats and aligning academic advances with industry needs.

Personality is a set of traits that describe people, such as emotions and behavioral patterns. Therefore, it is easy to believe that these patterns influence the way in which people interact with mobile devices. Since they are so widely used in everyday life and considering that most people have one of them, it becomes fundamental to create a dataset based on mobile device usage habits. The purpose of this is to recognize the personality of the users based on how they use their data. This paper presents the methodology in which the data was collected, as well as the neural networks used to perform preliminary personality recognition tests, using different metrics, getting promising personality prediction data with the use of ANN and MLP networks and the data obtained from mobile devices.

The Grouping Genetic Algorithm with Controlled Gene Transmission (GGA-CGT) is among the most effective algorithms for solving the one-dimensional Bin Packing Problem (1D-BPP), a classical NP-hard combinatorial optimisation problem with numerous industrial and logistical applications. This study aims to identify the characteristics that enable a mutation operator to perform better within this algorithm by implementing five state-of-the-art mutation operators: Elimination, Merge & Split, Swap, Insertion, and Item Elimination. Performance was evaluated in terms of the number of optimal solutions obtained. Our findings indicate that the GGA-CGT performs best with the least disruptive operators and that both the gene selection strategy and the item selection strategy can enhance the performance of mutation operators. These findings were validated by redesigning and improving a state-of-the-art item-oriented operator, achieving a 26% improvement over the best baseline version of the same operator.

Soil type is a critical factor influencing the seismic performance of buildings, as it affects the level of damage sustained during earthquakes. This paper presents a novel approach to classifying building soil types using pseudo-spectral acceleration readings recorded during seismic events. By leveraging machine learning classifiers, the study develops a model that accurately identifies soil types from pseudo-spectral acceleration data, achieving an accuracy of 89.16%. The methodology involves preprocessing the seismic data, extracting key features, and applying various classifiers to determine the most effective model. Performance is evaluated using metrics such as accuracy, precision, and recall. The findings indicate that this approach significantly improves soil classification accuracy over traditional methods, providing a practical tool for seismic hazard assessment and building design. This research further advances earthquake engineering by offering a data-driven solution to enhance building resilience.

Accurately modelling and predicting traffic congestion is essential for effective traffic management in dynamic urban environments. This study introduces a hybrid model that combines a deep neural network architecture, designed to capture long-term temporal dependencies, with physics-based approaches to traffic modelling. The deep learning component identifies complex temporal patterns and non-linear behaviours in traffic evolution, while the physics-based component incorporates dynamic constraints that enhance the model’s robustness and ability to generalise. This integration leverages the predictive power of machine learning while maintaining consistency with the fundamental principles governing vehicle flow. Experiments conducted using real-world traffic data demonstrate strong predictive performance, establishing a solid foundation for the development of intelligent transportation management systems and providing advanced tools to support adaptive mobility optimisation in congested urban settings.

The identification and characterisation of V genes poses a significant challenge due to the substantial number of alignments generated by diverse sequencing systems. This study proposes a method for the unification and reduction of the search space, with the objective of optimising the identification of V genes. The method integrates preprocessing, normalisation, and clustering using Gaussian Mixture Models. This approach facilitates data consolidation and reduces redundancy, thereby enhancing the efficiency and accuracy of the subsequent analysis. The elbow method was employed to determine the optimal number of groups, achieving a 98% reduction in the search space. The findings were validated through the use of metrics such as mean absolute error, mean squared error, and root mean squared error, thereby confirming the effectiveness of the method in improving the precision of gene identification.

Every year, 17.9 million people die from heart-failure–related conditions, making it the leading cause of mortality worldwide; early diagnosis could prevent many deaths. The most common way to detect cardiac abnormalities is through medical auscultation. Accurate diagnosis often depends on clinicians’ auscultation skills; however, they frequently have to listen in noisy environments. We propose a method for the automatic diagnosis of cardiovascular disease that is robust to noise. We extract entropy spectrograms from phonocardiograms to reliably convert audio signals into images, and then use a two-dimensional convolutional neural network (2D-CNN) to classify patients as healthy or unhealthy. To evaluate the method, we added white noise to the original recordings. The results show that entropy spectrograms are more robust than conventional feature-extraction techniques such as energy spectrograms or mel-frequency spectrograms.

In the environment of Big Data analytics worldwide, cloud web services were deployed across Internet and Intranet domains. Moreover, cloud computing, while possessing significant advantages and experiencing rapid development, faces trust complexities, privacy concerns, and security issues. These challenges necessitate the implementation of Quality of Service (QoS) measures in optimisation techniques for web service selection. This study focuses on the selection of component services and the use of an efficient algorithm with end-to-end quality measures. However, data diversification and service characteristics may reduce the accuracy of these measures. To address this, a novel QoS-based web service selection algorithm was developed, incorporating both weighted and subjective attributes. The proposed methodology employs a hybrid optimisation algorithm that integrates randomised attribute searches with the Invasive Weed Optimisation (IWO) algorithm. Furthermore, it calculates QoS measures based on the weighted attributes of web services. Many researchers have applied nature-inspired concepts to deal with optimisation complexities in Big Data, including the Eagle Perching Algorithm to improve the efficiency of cloud web services. In addition, the evolution of the Bald Eagle Search (BES) algorithm has been utilised as a nature-inspired approach, providing an efficient technique for optimisation problems by imitating the behaviour of bald eagles. The results demonstrate that the proposed methodology achieves improved performance metrics when compared with existing approaches, confirming its effectiveness in the evaluation of web service optimisation.

Quantum image processing represents a transformative approach to visual data analysis, leveraging the principles of quantum computing to overcome classical limitations. This work explores two prominent quantum image encoding methods: FRQI (Flexible Representation of Quantum Images) and NEQR (Novel Enhanced Quantum Representation). FRQI excels in qubit efficiency, making it suitable for hardware implementation, while NEQR offers superior precision in pixel intensity representation, ideal for complex image processing tasks. We detail the implementation of these algorithms, including preprocessing, quantum circuit design, and simulation, using platforms like Qiskit. The study highlights the potential of quantum image processing in fields such as medicine, industry, and environmental monitoring, while addressing challenges like qubit limitations and noise sensitivity. This research contributes to advancing quantum computing applications, paving the way for innovative and sustainable technological solutions.

Abstract. Indigenous languages like Purépecha face significant challenges in the modern era, particularly due to limited digital resources and a dwindling number of speakers. This study, conducted by researchers from CIC-IPN and CONACYT, presents an innovative application of transformer-based neural networks for the automatic translation of Purépecha to Spanish. Unlike previous works that utilized transformer architectures, this work develops a unique bilingual corpus through an algorithm based on the verbal inflection of Purépecha verbs, generating simple sentences in Purépecha and their corresponding Spanish translations. This corpus was then used to train a transformer model for automatic translation. The results indicate the potential of artificial intelligence to contribute to the preservation and revitalization of indigenous languages, opening new possibilities in the field of automatic translation and other natural language processing sectors.keywords in this section.