Optimization is an entire field that aims to improve efficiency and effectiveness across various domains. Its primary objective is to minimize costs, time, and risks while maximizing gains, quality, and efficiency. In this context, the 0/1 1-D bin packing problem is one of combinatorial optimization’s most challenging and extensively studied problems. This problem holds significant practical applications in supply chain management, packaging design, and resource optimization. This work solves the 0/1 1-D bin packing problem using a nature-inspired golden eagle optimizer. The hunting behavior of golden eagles inspires this bio-solver, and it employs swarm intelligence-based strategies to approximate solutions. We perform a comparative analysis of the bio-inspired algorithm to evidence its yield. We use twenty instances of the 1-D bin packing problem. Computational results show that the golden eagle optimizer exhibits better results in convergence time than well-known bio-inspired algorithms.

In some application fields, series are affected by two different types of effects: abrupt changes (or change-points) and functional effects. We propose here a Bayesian approach that allows us to estimate these two parts. Here, the underlying piecewise-constant part (associated to the abrupt changes) is expressed as the product of a lower triangular matrix by a sparse vector and the functional part as a linear combination of functions from a large dictionary where we want to select the relevant ones. This problem can thus lead to a global sparse estimation and a stochastic search variable selection approach is used to this end. The performance of our proposed method is assessed using simulation experiments. Applications to three real datasets from geodesy, agronomy and economy fields are also presented.

Exchange rates are determined by factors such as interest rates, political stability, confidence, the current account on balance of payments, government intervention, economic growth and relative inflation rates, among other variables. In October 2019, an increased climate of citizen discontent with current social policies resulted in a series of massive protests that ignited important political changes in Chile. This event along with the global COVID-19 pandemic were two major factors that affected the value of the US dollar and produced sudden changes in the typically stable USD/CLP (Chilean Peso) exchange rate. In this paper, we use a Bayesian approach to detect and locate change points in the currency exchange rate process in order to identify and relate these points with the important dates related to the events described above. The implemented method can successfully detect the onset of the social protests, the beginning of the COVID-19 pandemic in Chile and the economic reactivation in the US and Europe. In addition, we evaluate the performance of the proposed MCMC algorithms using a simulation study implemented in Python and R.

In complexity theory, there is a widely studied grouping of optimization problems that belongs to the non-deterministic polynomial-time hard set. One of them is the set covering problem, known as one of Karp’s 21 NP-complete problems, and it consists of finding a subset of decision variables for satisfying a set of constraints at the minimum feasible cost. However, due to the nature of the problem, this cannot be solved using traditional complete algorithms for hard instances. In this work, we present an improved binary version of the monkey search algorithm for solving the set covering problem. Originally, this approximate method was naturally inspired by the cognitive behavior of monkeys for climbing mountains. We propose a new climbing process with a better exploratory capability and a new cooperation procedure to reduce the number of unfeasible solutions. For testing this approach, we present a detailed computational results section, where we illustrate how this variation of the monkey search algorithm is capable of reaching various global optimums for a well-known instance set from the Beasley’s OR-Library and how it outperforms many other heuristics and meta-heuristics addressed in the literature. Moreover, we add a complete statistical analysis to show the effectiveness of the proposed approach with respect to the original version.

In this article we give a direct proof of a central limit theorem and law of large numbers for a functional with compact support of a diffusion. Some applications are given in order to obtain a parameter estimation for different models.

Emotion recognition through speech is a technique employed in various scenarios of Human–Computer Interaction (HCI). Existing approaches have achieved significant results; however, limitations persist, with the quantity and diversity of data being more notable when deep learning techniques are used. The lack of a standard in feature selection leads to continuous development and experimentation. Choosing and designing the appropriate network architecture constitutes another challenge. This study addresses the challenge of recognizing emotions in the human voice using deep learning techniques, proposing a comprehensive approach, and developing preprocessing and feature selection stages while constructing a dataset called EmoDSc as a result of combining several available databases. The synergy between spectral features and spectrogram images is investigated. Independently, the weighted accuracy obtained using only spectral features was 89%, while using only spectrogram images, the weighted accuracy reached 90%. These results, although surpassing previous research, highlight the strengths and limitations when operating in isolation. Based on this exploration, a neural network architecture composed of a CNN1D, a CNN2D, and an MLP that fuses spectral features and spectogram images is proposed. The model, supported by the unified dataset EmoDSc, demonstrates a remarkable accuracy of 96%.

Background Pain is modulated by expectation. Event-related potential (ERP) studies of the influence of expectation on pain typically utilise laser heat stimulation to provide a controllable nociceptive-specific stimulus. Painful electric stimulation has a number of practical advantages, but is less nociceptive-specific. We compared the modulation of electric versus laser-evoked pain by expectation, and their corresponding pain-evoked and anticipatory ERPs. New method We developed understanding of recognised methods of laser and electric stimulation. We tested whether pain perception and neural activity induced by electric stimulation was modulated by expectation, whether this expectation elicited anticipatory neural correlates, and how these measures compared to those associated with laser stimulation by eliciting cue-evoked expectations of high and low pain in a within-participant design. Results Despite sensory and affective differences between laser and electric pain, intensity ratings and pain-evoked potentials were modulated equivalently by expectation, though ERPs only correlated with pain ratings in the laser pain condition. Anticipatory correlates differentiated pain intensity expectation to laser but not electric pain. Comparison with existing method Previous studies show that laser-evoked potentials are modulated by expectation. We extend this by showing electric pain-evoked potentials are equally modulated by expectation, within the same participants. We also show a difference between the pain types in anticipation. Conclusions Though laser-evoked potentials express a stronger relationship with pain perception, both laser and electric stimulation may be used to study the modulation of pain-evoked potentials by expectation. Anticipatory-evoked potentials are elicited by both pain types, but they may reflect different processes.

We propose an efficient computational method to obtain the fractional derivative of a digital signal. The proposal consists of a new interpretation of the Grünwald–Letnikov differintegral operator where we have introduced a finite Cauchy convolution with the Grünwald–Letnikov dynamic kernel. The method can be applied to any signal without knowing its analytical form. In the experiments, we have compared the proposed Grünwald–Letnikov computational fractional derivative method with the Riemman–Louville fractional derivative approach for two well-known functions. The simulations exhibit similar results for both methods; however, the Grünwald–Letnikov method outperforms the other approach in execution time. Finally, we show an application of how our proposal can be useful to find the fractional relationship between two well-known biomedical signals.

Mammography is known as one of the best forms to screen possible breast cancer in women, and recently deep learning models have been developed to assist the radiologist in the diagnosis. However, their lack of interpretability has become a significant drawback to their extended use in clinical practice. This paper introduces a novel approach for detecting and localising pathological findings in mammography exams through the use of a EfficientNet-based deep learning model. The model is trained using cropped segments of labelled pathological findings from Vindr Mammography Dataset. Achieving an average F1-score of 72.7 %, and reaching on mass and suspicious calcifications an F1-Score of 79.9 % and 84.5 % respectively. Using this classifier we propose a method to visualise from local information the regions of interest where pathological findings could be present on the complete image. Plus, we describe the limitations regarding area coverage of these patches on the model's capability of generalization and certainty on its predictions, explaining its functionality.

The influence spread in a social network is an iterative process that can take several steps. It begins with an activation seed and finishes when the current activation cannot influence more actors. The multi-objective influence spread problem corresponds to finding the smallest number of actors capable of maximizing the influence spread within the network. This problem has been solved by metaheuristic optimization algorithms using swarm intelligence methods. This article proposes a heuristic to improve the existing solution: when two sets of actors can influence the same number of actors, the one whose spread requires the least number of steps is chosen. The proposed solution is tested on two different real networks. The results show that the heuristic allowed better results for both networks and decreased the average number of steps in the influence spread processes (in 15.5 and 0.07 average steps, respectively), thus improving execution times. Moreover, the heuristic allowed decreasing the number of steps in 83% (against 17% of increasing) and 13% (against 7% of increasing) of the particles, respectively.

Sex differences have been a rarely addressed aspect in digital game-based learning (DGBL). Likewise, mixed results have been presented regarding the effects according to sex and the conditions that generate these effects. The present work studied the effects of a drill-and-practice mathematical game on primary students. The study focused on an analysis by sex, measuring motivation and learning in the practice activity. Also, two instructional mechanics were considered regarding the question answering to search for possible differences: a multiple-try feedback (MTF) condition and a single-try feedback (STF) condition. A total of 81 students from four courses and two schools participated in the intervention. The study’s main findings were as follows: (a) the girls outperformed the boys in terms of the students’ learning gains; (b) the girls presented lower levels of competence and autonomy than the boys; (c) under MTF, the girls presented lower levels of autonomy but no differences in competence contrasted with the boys; (d) under STF, the girls presented lower levels of competence but no differences in autonomy contrasted with the boys; (e) no sex differences existed in interest, effort, and value, in general, as per the instructional condition. This study enhances the knowledge of sex differences under diverse instructional settings, in particular providing insights into the possible differences by sex when varying the number of attempts provided to students.

Cryptocurrencies have been receiving the sustained attention of investors since 2009. These new investment vehicles are digitally native, meaning that they are traded exclusively on 24/7 digital platforms. Consequently, they offer an excellent scenario to test the Efficient Market Hypothesis, by developing algorithm-based trading strategies. Such strategies aim to beat the market. It has been previously reported that daily returns do not exhibit long range dependence. However, daily volatility in major cryptocurrencies is highly persistent. Therefore, buy/hold/sell decision support systems could be able to capture such market inefficiency. This is especially important for investors interested in periodically trading a set of cryptocurrencies, in order to maximize their wealth. This paper presents a dynamic linguistic decision making approach for building decision models to support cryptocurrency investors in buy/hold/sell decisions. This approach exhibits a good computational performance for obtaining recommendations based on quantitative data. Moreover, this procedure is able to identify some inefficient cryptocurrency behaviors which are not captured by traditional econometric techniques. Our results uncover arbitrage opportunities that outperform buy-and-hold or random strategies.

Within the e-commerce sphere, optimizing the product classification process assumes pivotal importance, owing to its direct influence on operational efficiency and profitability. In this context, employing machine learning algorithms stands out as a premier solution for effectively automating this process. The design of these models commonly adopts either a flat or local (hierarchical) approach. However, each of them exhibits significant limitations. The regional approach introduces taxonomic inconsistencies in predictions, whereas the flat approach becomes inefficient when dealing with extensive datasets featuring high granularity. Therefore, our research introduces a solution for hierarchical product classification based on a Machine Learning model that integrates flat and local (hierarchical) classification approaches using a 4-level electronic product dataset obtained from a renowned e-commerce platform in Latin America. In pursuit of this goal, a comparative analysis of seven machine learning algorithms, including Multinomial Naive Bayes, Linear Support Vector Classifier, Multinomial Logistic Regression, Random Forest, XGBoost, FastText, and Voting Ensemble, was conducted. This hybrid approach model performs better than models using a single approach. It surpassed the top-performing flat approach model by 0.15% and outperformed the leading local approach (Local Classifier per Level) model by 4.88%, as measured by the weighted F1-score. Additionally, this paper contributes to the academic community by presenting a significant Spanish-language dataset comprising over one million products and discussing the preprocessing techniques tailored for the dataset. It also addresses the study's inherent limitations and potential avenues for future exploration in this field.

Multimodal methods for emotion recognition consider several sources of data to predict emotions; thus, a fusion method is needed to aggregate the individual results. In the literature, there is a high variety of fusion methods to perform this task, but they are not suitable for all scenarios. In particular, there are two relevant aspects that can vary from one application to another: (i) in many scenarios, individual modalities can have different levels of data quality or even be absent, which demands fusion methods able to discriminate non-useful from relevant data; and (ii) in many applications, there are hardware restrictions that limit the use of complex fusion methods (e.g., a deep learning model), which could be quite computationally intensive. In this context, developers and researchers need metrics, guidelines, and a systematic process to evaluate and compare different fusion methods that can fit to their particular application scenarios. As a response to this need, this paper presents a framework that establishes a base to perform a comparative evaluation of fusion methods to demonstrate how they adapt to the quality differences of individual modalities and to evaluate their performance. The framework provides equivalent conditions to perform a fair assessment of fusion methods. Based on this framework, we evaluate several fusion methods for multimodal emotion recognition. Results demonstrate that for the architecture and dataset selected, the methods that best fit are: Self-Attention and Weighted methods for all available modalities, and Self-Attention and Embracenet+ when a modality is missing. Concerning the time, the best times correspond to Multilayer Perceptron (MLP) and Self-Attention models, due to their small number of operations. Thus, the proposed framework provides insights for researchers in this area to identify which fusion methods better fit their requirements, and thus to justify the selection.

This paper focuses on modeling and solving the Manufacturing Cell Design Problem (MCDP) by using the Harmony Search (HS) metaheuristic. The MDCP consists on grouping machines and parts that they process, into groups called cells. So, the idea is to identify an organization of cells such that the number of times that a piece is transported between these cells is minimized. To this end, we use the HS optimization algorithm, which is based on the process of improvisation performed by musicians to find a perfect musical harmony. The experimental results demonstrate the efficiency of the proposed approach which is able to reach all global optimums for a set of 90 well-known MDCP instances.

Visually evoked steady-state potentials (SSVEPs) are neural responses elicited by visual stimuli oscillating at specific frequencies. In this study, we introduce a novel LED stimulator system explicitly designed for steady-state visual stimulation, offering precise control over visual stimulus parameters, including frequency resolution, luminance, and the ability to control the phase at the end of the stimulation. The LED stimulator provides a personalized, modular, and affordable option for experimental setups. Based on the Teensy 3.2 board, the stimulator utilizes direct digital synthesis and pulse width modulation techniques to control the LEDs. We validated its performance through four experiments: the first two measured LED light intensities directly, while the last two assessed the stimulator’s impact on EEG recordings. The results demonstrate that the stimulator can deliver a stimulus suitable for generating SSVEPs with the desired frequency and phase resolution. As an open source resource, we provide comprehensive documentation, including all necessary codes and electrical diagrams, which facilitates the system’s replication and adaptation for specific experimental requirements, enhancing its potential for widespread use in the field of neuroscience setups.

Pregnancy in-vitro fertilization (IVF) cases are associated with adverse first-trimester outcomes in comparison to spontaneously achieved pregnancies. Human chorionic gonadotrophin β subunit (β-HCG) is a well-known biomarker for the diagnosis and monitoring of pregnancy after IVF. Low levels of β-HCG during this period are related to miscarriage, ectopic pregnancy, and IVF procedure failures. Longitudinal profiles of β-HCG can be used to distinguish between normal and abnormal pregnancies and to assist and guide the clinician in better management and monitoring of post-IVF pregnancies. Therefore, assessing the association between longitudinally measured β-HCG serum concentration and time to early miscarriage is of crucial interest to clinicians. A common joint modeling approach is to use the longitudinal β-HCG trajectory to determine the risk of miscarriage. This work was motivated by a follow-up study with normal and abnormal pregnancies where β-HCG serum concentrations were measured in 173 young women during a gestational age of 9-86 days in Santiago, Chile. Some women experienced a miscarriage event, and their exact event times were unknown, so we have interval-censored data, with the event occurring between the last time of the observed measurement and ten days later. However, for those women belonging to the normal pregnancy group; that is, carrying a pregnancy to a full-term event, right censoring data are observed. Estimation procedures are based on the Stochastic Approximation of the Expectation-Maximization (SAEM) algorithm.

Bio-inspired computing is an engaging area of artificial intelligence which studies how natural phenomena provide a rich source of inspiration in the design of smart procedures able to become powerful algorithms. Many of these procedures have been successfully used in classification, prediction, and optimization problems. Swarm intelligence methods are a kind of bio-inspired algorithm that have been shown to be impressive optimization solvers for a long time. However, for these algorithms to reach their maximum performance, the proper setting of the initial parameters by an expert user is required. This task is extremely comprehensive and it must be done in a previous phase of the search process. Different online methods have been developed to support swarm intelligence techniques, however, this issue remains an open challenge. In this paper, we propose a hybrid approach that allows adjusting the parameters based on a state deducted by the swarm intelligence algorithm. The state deduction is determined by the classification of a chain of observations using the hidden Markov model. The results show that our proposal exhibits good performance compared to the original version.

Contemporary education is a vast field that is concerned with the performance of education systems. In a formal e-learning context, student dropout is considered one of the main problems and has received much attention from the learning analytics research community, which has reported several approaches to the development of models for the early prediction of at-risk students. However, maximizing the results obtained by predictions is a considerable challenge. In this work, we developed a solution using only students’ interactions with the virtual learning environment and its derivative features for early predict at-risk students in a Brazilian distance technical high school course that is 103 weeks in duration. To maximize results, we developed an elitist genetic algorithm based on Darwin’s theory of natural selection for hyperparameter tuning. With the application of the proposed technique, we predicted the student at risk with an Area Under the Receiver Operating Characteristic Curve (AUROC) above 0.75 in the initial weeks of a course. The results demonstrate the viability of applying interaction count and derivative features to generate prediction models in contexts where access to demographic data is restricted. The application of a genetic algorithm to the tuning of hyperparameters classifiers can increase their performance in comparison with other techniques.