This paper presents a systematic literature review of machine learning models used for predicting Bitcoin prices. We identified and analyzed 26 different machine learning models notable for their effectiveness in predictive tasks. Out of these, 6 models are discussed in detail, focusing on their advantages, disadvantages, and potential hybrid approaches. Additionally, we collected 21 evaluation metrics, identifying 8 as the most relevant for cryptocurrency price prediction. In terms of datasets, we found a reliance on public sources such as Kaggle and Yahoo Finance; however, challenges related to the inconsistency in data availability remain. Lastly, we noted a lack of standardized procedures for comparing models, highlighting the need for the development of systematic methodologies to standardize evaluations in this field. Smart citations: https://scite.ai/reports/10.61467/2007.1558.2026.v17i2.1166Dimensions.Open Alex.

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.

Didactic control systems are tools that facilitate the understanding of topics related to control and stabilisation. However, due to the high cost of these devices, their implementation in schools is limited. This work presents the design, construction, and testing of a low-cost didactic prototype of a Furuta-type pendulum for the analysis of topics related to this system. The construction included parts designed in SolidWorks and produced with 3D printing, threaded rods, a precision linear potentiometer, a DC motor, a BTS7960 H-bridge, and an Arduino Nano. The functionality of the project was verified through the implementation of two digital tools: a Kalman filter to improve the measured signal and a PID controller to regulate the angular speed of the motor. It was demonstrated that the prototype can be stabilised under the operational considerations of PID control without the use of an angular position sensor. Moreover, the Kalman filter must be kept small in order to avoid information loss.

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.

An automatic temperature control system is an important application used in almost all modern gadgets, smart homes and industries; however, there are almost no available prototypes to practise with, and the ones that do exist cost more than an average student in Mexico can afford. This project is based on the implementation of a digital PID controller, realised through widely studied mathematical concepts, applied to a fan that blows air into an enclosed area housing an incandescent bulb, thereby controlling the internal temperature. A Kalman filter was used to reduce the thermocouple noise. The performance of the controller was tested by observing its ability to maintain the desired temperature setpoint steadily. This test highlighted the PID controller’s ability to reach and maintain the desired temperature, validating its effectiveness as both an educational tool and a method of control for industrial applications.

This paper presents a systematic literature review of the application of retrieval-augmented generation (RAG) systems in educational settings, with a focus on teaching software engineering and related computing disciplines. Drawing on case studies, academic experiments, and surveys of teachers and students, it provides an overview of the current landscape, highlighting perceptions, reported effectiveness, and the technology’s impact in academia. Based on an analysis of 71 selected scientific papers, the review synthesises evidence on the extent to which RAG systems mitigate hallucinations and improve human–AI interaction. In addition, it suggests that many approaches discussed across studies could be strategically aligned with the integration of DevOps practices and RAG, enhancing their use through automation, continuous improvement, and the agile adoption of technologies within educational processes.

Cloud computing is essential for executing scientific workflows, offering scalable resources to process large volumes of data. However, efficient resource allocation remains a challenge due to fluctuations in demand and inadequate planning, which can increase makespan and thus energy consumption. In this context, infrastructure as a service is primarily used, allowing the rental of virtual machines with different characteristics. This study focuses on identifying the factors that influence the improvement of makespan for executing workflows from different disciplines and characteristics. The factors evaluated in this study are workfloF size, structure, and the number of virtual machines (VMs) required. The algorithms in this study consist of four heuristics and one metaheuristic, specifically a genetic algorithm (GA). Regarding the factors, the most relevant in CC is the number of VMs, where increasing the number of VMs reduces makespan to a certain extent, following the behavior described by the law of diminishing returns.

Incomplete data poses a significant obstacle in data science and machine learning, influencing model outcomes and occurring commonly across domains such as health, nutrition, electricity, agriculture, chemistry and water resources. Missing data refers to the absence of information for one or more variables in a dataset. Accurate imputation is therefore crucial to ensure the reliability and validity of analyses and predictive models. This study proposes a Harmony Search Algorithm (HSA) to address missing-data imputation, emphasising its flexibility and adaptability. The approach seeks the best imputations by minimising error metrics such as MAE, MSE and RMSE. Computational tests indicate that HSA is a promising method for imputing missing data in a range of contexts.

In this contribution, the Markov chain approach is used to predict the job satisfaction of employees in a telecommunications company in Mexico. Under certain assumptions, the Chapman–Kolmogorov equation is applied to estimate the probabilities that: employees remain satisfied if they are satisfied; if they are not satisfied, the probability that they remain unsatisfied; if they are satisfied, the probability that they will become unsatisfied; and finally, if they are unsatisfied, the probability that they will become satisfied, within a one-year prediction horizon. Our proposal could be applied to support decision-making in human capital management strategies.

This study investigates the performance of the YOLOv8 object detection algorithm for fuel tank detection using unmanned aerial vehicles (UAVs) under various conditions and perspectives. UAVs, with their agile mobility and ability to capture high-quality images from different angles, are proven to facilitate the detection of fuel tanks, which play a critical role in various industries. The integration of UAVs with YOLOv8 has led to significant advancements in automation and precision in inspection processes within the energy and industrial sectors. YOLOv8, a single-stage deep learning-based object detection model, demonstrates superior architecture compared to previous YOLO iterations, showcasing its strengths in speed and accuracy for object detection tasks. The model achieves a precision of 0.888, recall of 0.896, and mAP of 0.891, confirming its strong capabilities in detecting fuel tanks and supporting the sustainability of industrial and energy operations. With a processing time of 41 ms, YOLOv8 proves to be highly effective for real-time applications. This research highlights the importance of optimizing UAVs and deep learning models for reliable data collection in challenging environments and demonstrates their potential for use in fuel tank monitoring and infrastructure reliability tracking across industries. While accurate detection successes are noted, the study emphasizes the need for further optimization of the algorithm to address false positives and undetected objects in real-world applications. Future work will explore the adaptation of high-performance algorithms such as YOLOv8 for broader object detection scenarios and their testing under diverse environmental conditions.