Abstract: The identification of seeds from native species is a complex assessment due to the high Brazilian biodiversity and varied characteristics between species. The objective was to apply different machine learning classifiers associated with image analysis to identify seeds of forest species. In total, 155 native species belonging to 42 botanical families were analyzed. In addition, to determine the appropriate machine learning classifier, five supervised learning classification techniques were implemented: decision trees (DT), artificial neural networks (ANN), k-nearest neighbors (k-NN), Naive-Bayes classifier (NBC) and support vector machine (SVM), which had their performance evaluated. For modeling, 66% of the seeds’ morphobiometric data were used to train the classifiers, while 34% were reserved for validation. The classifiers are promising tools for identifying species from seed images. The decision tree (DT) classifier showed greater accuracy for correct species identification (82.8%), followed by ANN (81.7%), k-NN (81.7%), NBC (81.1%) and SVM (78.7%). Therefore, it is possible to identify seeds of native species from images and machine learning with a satisfactory accuracy rate. Finally, the decision tree classifier is recommended.

Resumo: A identificação de sementes de espécies nativas é uma avaliação complexa devido a elevada biodiversidade brasileira e características variadas entre as espécies. Objetivou-se aplicar diferentes classificadores de aprendizado de máquina associado à análise de imagens para identificar sementes de espécies florestais. Foram analisadas 155 espécies nativas pertencentes a 42 famílias botânicas. Para determinar o classificador de aprendizado de máquina adequado, cinco técnicas de classificação por aprendizado supervisionado foram implementadas: árvores de decisão (DT), redes neurais artificiais (ANN), k-vizinhos mais próximos (k-NN), classificador Naive-Bayes (NBC) e máquina de vetores de suporte (SVM), os quais tiveram seu desempenho avaliados. Para modelagem, 66% dos dados morfobiométricos das sementes foram usados para treinamento dos classificadores, enquanto 34% foram reservados para validação. Os classificadores são ferramentas promissoras para a identificação de espécies a partir das imagens de sementes. O classificador por árvores de decisão (DT) apresentou maior acurácia para identificação correta das espécies (82,8%), seguido dos classificadores ANN (81,7%), k-NN (81,7%), NBC (81,1%) e SVM (78,7%). Portanto, é possível realizar a identificação de sementes de espécies nativas a partir de imagens e aprendizado de máquina com taxa satisfatória de acurácia. Recomenda-se o classificador por árvores de decisão.

Con el avance de la tecnología, el aprendizaje en teleconsultas pasó a formar parte de la graduación médica actual. Esta investigación presenta las representaciones de un grupo de alumnos de un curso de medicina localizado en la capital del estado de São Paulo sobre sus experiencias con la realización de teleconsultas sincronizadas supervisadas durante una pasantía realizada en un ambulatorio escuela, durante el segundo trimestre de 2021. Se traba de una investigación cualitativa que utilizó el software IRaMuTeQ® y Análisis de Contenido. Estos alumnos puntúan obstáculos para la utilización de las teleconsultas, destacando preocupaciones con la humanización, seguridad de la persona atendida y calidad de la consulta, pero visualizan potencialidades tales como la facilidad de acceso y la comodidad. Reconocen la importancia de la práctica durante su graduación en un mundo cada vez más tecnológico, pero presentan inquietudes relacionadas a la pérdida del examen físico y a la mercantilización de la medicina.

Com o avançar da tecnologia, passou a fazer parte da graduação médica atual o aprendizado em teleconsultas. Esta pesquisa apresenta as representações de um grupo de alunos de um curso de Medicina localizado na capital do estado de São Paulo sobre suas experiências com a realização de teleconsultas síncronas supervisionadas durante estágio realizado em um ambulatório-escola, durante o segundo semestre de 2021. Trata-se de pesquisa qualitativa que utilizou o software IRaMuTeQ® e “Análise do Conteúdo”. Esses educandos pontuam entraves para a utilização das teleconsultas, destacando preocupações com a humanização, segurança da pessoa atendida e qualidade da consulta, mas visualizam potencialidades tais como a facilidade de acesso e a comodidade. Reconhecem a importância da prática durante sua Graduação em um mundo cada vez mais tecnológico, mas trazem inquietações relacionadas à perda do exame físico e à mercantilização da Medicina.

With the progress of technology, teleconsultation has become part of medical degree courses. This article presents the representations of a group of medical students from a university in the capital of the state of São Paulo regarding their experiences with supervised synchronous teleconsultations during an internship in a teaching clinic in the second semester of 2021. We conducted a qualitative study using the software IRaMuTeQ® and content analysis. The students highlighted barriers to the use of teleconsultation, emphasizing concerns with humanization, patient safety and consultation quality, but also perceived strengths such as ease of access and convenience. They recognize the importance of practice on the degree course in an ever more technological world, but raise concerns about the lack of physical examination and the commodification of medicine.

RESUMEN Introducción: la sintomatología depresiva es altamente prevalente en la población peruana. El uso del algoritmo de árbol de decisiones podría beneficiar en hallar grupos especialmente vulnerables a padecer síntomas depresivos. Objetivo: determinar los grupos especialmente vulnerables a tener síntomas depresivos según factores sociodemográficos mediante algoritmo de árbol de decisiones por aprendizaje automático. Material y métodos: se aplicó un diseño observacional, descriptivo, retrospectivo y transversal. Los datos provinieron de la encuesta nacional demográfica y de salud. La población fue 32.062 adultos. La variable dependiente fue: presencia de síntomas depresivos, y como variables explicativas: grupo etario, lengua materna, grupo étnico, nivel educativo, edad de inicio de consumo de alcohol, consumo de alcohol, estado conyugal, sexo. Se utilizó el algoritmo de árbol de decisiones mediante detección automática de interacciones mediante chi-cuadrado (CHAID). Resultados: las variables significativas en el algoritmo fueron: sexo, tipo de lengua materna, estado conyugal, grupo etario, nivel educativo alcanzado, clasificando de forma correcta 75,80% de los casos de síntomas depresivos. Los nodos asociados principalmente a la presencia de síntomas depresivos fueron: nodo 2 (sexo femenino), nodo 6 (adultos desde 39 años), nodo 13 (educación hasta secundaria). Según sexo, en mujeres, las variables principalmente asociadas fueron los correspondientes al nodo 2 (adultos desde los 39 años), nodo 5 (educación hasta secundaria) y nodo 13 (lengua materna originaria). En hombres, los nodos asociados principalmente a síntomas depresivos fueron el nodo 2 (lengua materna originaria), nodo 6 (adultos desde los 39 años) y nodo 11 (nivel educativo alcanzado hasta secundaria). Conclusiones: el principal grupo sociodemográfico asociado al desarrollo de síntomas depresivos son el sexo femenino, desde los 39 años y cuya educación ha llegado a la etapa escolar. El uso de algoritmos de aprendizaje automático es útil para crear herramientas de cribado de poblaciones vulnerables a padecer síntomas depresivos.

ABSTRACT Introduction: Depressive symptoms are highly prevalent in the Peruvian population. The use of the decision tree algorithm could be beneficial in finding groups especially vulnerable to suffering from depressive symptoms. Objective: To determine the groups especially vulnerable to having depressive symptoms according to sociodemographic factors using a machine learning decision tree algorithm. Material and methods: An observational, descriptive, retrospective and cross-sectional design was applied. Data came from the National Demographic and Health Survey. The population was 32,062 adults and the dependent variable was the presence of depressive symptoms, and as explanatory variables: age group, mother tongue, ethnic group, educational level, age of onset of alcohol consumption, alcohol consumption, marital status, sex. The decision tree algorithm using automatic chi-square interaction detection (CHAID) was used. Results: The significant variables in the algorithm were sex, type of mother tongue, marital status, age group, educational level achieved, correctly classifying 75.80% of the cases of depressive symptoms. The nodes mainly associated with the presence of depressive symptoms were: node 2 (female sex), node 6 (adults from 39 years old), and node 13 (education up to secondary school). According to sex, in women, the variables mainly associated were those corresponding to node 2 (adults from 39 years of age), node 5 (education up to secondary school) and node 13 (original mother tongue). In men, the nodes mainly associated with depressive symptoms were node 2 (native mother tongue), node 6 (adults from 39 years of age) and node 11 (educational level reached up to secondary school). Conclusions: The main sociodemographic group associated with the development of depressive symptoms is the female sex, from the age of 39 and whose education has reached the school stage. The use of machine learning algorithms is useful to create screening tools for populations vulnerable to suffering from depressive symptoms.

ABSTRACT Introduction: Depressive symptoms are highly prevalent in the Peruvian population. The use of the decision tree algorithm could be beneficial in finding groups especially vulnerable to suffering from depressive symptoms. Objective: To determine the groups especially vulnerable to having depressive symptoms according to sociodemographic factors using a machine learning decision tree algorithm. Material and methods: An observational, descriptive, retrospective and cross-sectional design was applied. Data came from the National Demographic and Health Survey. The population was 32,062 adults and the dependent variable was the presence of depressive symptoms, and as explanatory variables: age group, mother tongue, ethnic group, educational level, age of onset of alcohol consumption, alcohol consumption, marital status, sex. The decision tree algorithm using automatic chi-square interaction detection (CHAID) was used. Results: The significant variables in the algorithm were sex, type of mother tongue, marital status, age group, educational level achieved, correctly classifying 75.80% of the cases of depressive symptoms. The nodes mainly associated with the presence of depressive symptoms were: node 2 (female sex), node 6 (adults from 39 years old), and node 13 (education up to secondary school). According to sex, in women, the variables mainly associated were those corresponding to node 2 (adults from 39 years of age), node 5 (education up to secondary school) and node 13 (original mother tongue). In men, the nodes mainly associated with depressive symptoms were node 2 (native mother tongue), node 6 (adults from 39 years of age) and node 11 (educational level reached up to secondary school). Conclusions: The main sociodemographic group associated with the development of depressive symptoms is the female sex, from the age of 39 and whose education has reached the school stage. The use of machine learning algorithms is useful to create screening tools for populations vulnerable to suffering from depressive symptoms.

RESUMEN Introducción: la sintomatología depresiva es altamente prevalente en la población peruana. El uso del algoritmo de árbol de decisiones podría beneficiar en hallar grupos especialmente vulnerables a padecer síntomas depresivos. Objetivo: determinar los grupos especialmente vulnerables a tener síntomas depresivos según factores sociodemográficos mediante algoritmo de árbol de decisiones por aprendizaje automático. Material y métodos: se aplicó un diseño observacional, descriptivo, retrospectivo y transversal. Los datos provinieron de la encuesta nacional demográfica y de salud. La población fue 32.062 adultos. La variable dependiente fue: presencia de síntomas depresivos, y como variables explicativas: grupo etario, lengua materna, grupo étnico, nivel educativo, edad de inicio de consumo de alcohol, consumo de alcohol, estado conyugal, sexo. Se utilizó el algoritmo de árbol de decisiones mediante detección automática de interacciones mediante chi-cuadrado (CHAID). Resultados: las variables significativas en el algoritmo fueron: sexo, tipo de lengua materna, estado conyugal, grupo etario, nivel educativo alcanzado, clasificando de forma correcta 75,80% de los casos de síntomas depresivos. Los nodos asociados principalmente a la presencia de síntomas depresivos fueron: nodo 2 (sexo femenino), nodo 6 (adultos desde 39 años), nodo 13 (educación hasta secundaria). Según sexo, en mujeres, las variables principalmente asociadas fueron los correspondientes al nodo 2 (adultos desde los 39 años), nodo 5 (educación hasta secundaria) y nodo 13 (lengua materna originaria). En hombres, los nodos asociados principalmente a síntomas depresivos fueron el nodo 2 (lengua materna originaria), nodo 6 (adultos desde los 39 años) y nodo 11 (nivel educativo alcanzado hasta secundaria). Conclusiones: el principal grupo sociodemográfico asociado al desarrollo de síntomas depresivos son el sexo femenino, desde los 39 años y cuya educación ha llegado a la etapa escolar. El uso de algoritmos de aprendizaje automático es útil para crear herramientas de cribado de poblaciones vulnerables a padecer síntomas depresivos.

Resumo As práticas pedagógicas supervisionadas apresentam um grande potencial para melhorar a qualidade dos processos de ensino e aprendizagem. Considerando as amplas possibilidades que oferecem, o presente trabalho tem como finalidade analisar as práticas pedagógicas supervisionadas de várias estudantes inscritas no Mestrado em Ensino do 1º e do 2º CEB, na vertente de História e Geografia de Portugal, implementado na Escola Superior de Educação do Instituto Politécnico de Lisboa (ESELx). Procura, ainda, avaliar as diferentes fases que compõem o ciclo da supervisão, tendo em conta o papel relevante de cada uma delas na avaliação da prática docente. A metodologia de recolha e análise de dados é de natureza qualitativa, utilizando a observação não participante nos contextos de sala de aula, e a análise de conteúdo de vários documentos, entre os quais se destaca o Plano de Intervenção Educativa concebido pelas estudantes. Entre os resultados, reconhece-se a relevância de criar espaços contínuos de reflexão e diálogo com o estudante/estagiário, os quais favorecem o questionamento e a melhoria, tanto da atividade do supervisor como das competências do futuro professor. Por outro lado, identificam-se possibilidades de melhoria em relação à mobilização de competências específicas da educação histórica e geográfica. Mais concretamente, no que diz respeito ao desenvolvimento de estratégias de questionamento e problematização da realidade, que promovam análises críticas e reflexivas, orientadas para a sua transformação.

Résumé Les pratiques pédagogiques supervisées ont un grand potentiel pour améliorer la qualité des processus d'enseignement et d'apprentissage. Compte tenu des vastes possibilités qu'ils offrent, le présent travail vise à analyser les pratiques pédagogiques supervisées de plusieurs étudiants du Master en enseignement des 1er et 2ème CEB, dans le domaine de l'histoire et de la géographie du Portugal, à l'École Supérieure d'Éducation de l'Institut Polytechnique de Lisbonne (ESELx). Plus, il vise à évaluer les différentes phases qui composent le cycle de supervision, et mis en évidence le rôle pertinent de chacune d’elles dans l'évaluation de la pratique enseignante. A cette fin, nous avons utilisé une méthodologie qualitative, en recourant à l'observation non participante dans des contextes de classe, et l'analyse de contenu de divers documents, en mettant l’accent sur le projet d'intervention éducative, conçu par chaque étudiant. Parmi les résultats, la pertinence de créer des espaces continus de réflexion et de dialogue avec l’étudient est reconnue, ce qui favorise le questionnement et l'amélioration tant de l'activité du superviseur que des compétences des futurs maitres. D'autre part, il y a des possibilités d'amélioration en ce qui concerne la mobilisation des compétences spécifiques de l'enseignement de l'histoire et de la géographie. Plus précisément, en ce qui concerne le développement de stratégies de questionnement et de problématisation de la réalité, qui favorisent les analyses critiques et réflexives visant à leur transformation.

Resumen Las prácticas pedagógicas supervisadas poseen un gran potencial para mejorar la calidad los procesos de enseñanza y aprendizaje. Considerando las amplias oportunidades que ofrece, el presente trabajo tiene como finalidad analizar las prácticas pedagógicas supervisadas de varias alumnas adscritas al Máster de Historia y Geografía de Portugal en el 2º Ciclo de Educación Básica, impartido en la Escola Superior de Educação del Instituto Politécnico de Lisboa (ESELx). Asimismo, persigue valorar las distintas fases que componen el ciclo de supervisión, considerando el papel relevante de cada una de ellas en la evaluación del ejercicio docente. Para ello, se emplea una metodología cualitativa, utilizando la observación no participante en el contexto de aula y el análisis de contenido de varios documentos, entre los que destaca el Plan de Intervención Educativa diseñado por los estudiantes. Entre los resultados, se reconoce la relevancia de crear espacios continuos de reflexión y diálogo con el practicante, los cuales favorecen el cuestionamiento y el perfeccionamiento tanto de las labores del supervisor como de las competencias del aprendiz. Por otro lado, se manifiestan posibilidades de mejora en relación con la movilización de competencias específicas de educación histórica y geográfica. Concretamente, en lo relativo al desarrollo de estrategias de cuestionamiento y problematización de la realidad, que promuevan análisis críticos y reflexivos orientados a su transformación.

Abstract The supervised teaching practices Supervised Pedagogical Practices have great potential to improve the quality of teaching and learning processes. Considering the wide possibilities they offer, the present work aims to analyse the supervised pedagogical practices of several students of the Master's in Teaching of the 1st and 2nd CEB, in the area of History and Geography of Portugal, at the Lisbon Higher School of Education of the Polytechnic Institute of Lisbon (ESELx). In addition, it seeks to assess the different phases that make up the supervision cycle, taking into account the relevant role of each of them in the evaluation of teaching practice. For this, we employed a qualitative methodology, using non-participant observation in classroom contexts, and content analysis of various documents, with emphasis on the Educational Intervention Project, designed by each student. Among the results, the relevance of creating continuous spaces for reflection and dialogue with the trainee is recognized, which favors the questioning and improvement of both the supervisor's activity and the student's skills. On the other hand, there are possibilities for improvement in what concerns the mobilization of specific competences of historical and geographical education. More specifically, with regard to the development of strategies for questioning and problematizing reality, which promote critical and reflective analysis, oriented towards change.

Abstract: The COVID-19 outbreak commenced in Wuhan, China, in December 2019 and swiftly disseminated worldwide. On March 11, 2020, the World Health Organization (WHO) formally designated the COVID-19 outbreak as a global pandemic [1]. This highly contagious disease, which has also started to spread among young people, necessitates the implementation of policies to avert the collapse of hospitals due to shortages in beds, mechanical ventilators, and intensive care units. Using the record from March 2020 to May 31 2021, (before the arrival of vaccines in Mexico) of the website of the General Directorate of Epidemiology of the Ministry of Health of the Government of the Mexican Republic, with more than seven million people associated with COVID-19, a model is built with twelve characteristics that predicts the risk of death associated with COVID-19, applying Supervised Learning Algorithms such as Linear Regression, Naive Bayes, Decision Trees and Random Forests. The results of the machine learning algorithms show a performance of 87%. Subsequently, the model was tested again with 402,116 COVID-19-associated patients from the month of June, achieving an accuracy of 91%, surpassing the model proposed in [3]. When performing data cleaning, we observed that the twelve variables are not correlated, unlike what the authors showed in [4], where data cleaning was not carried out. This analysis can aid in designing strategies and policies to combat its spread and prevent mortality, as well as assist hospitals in prioritizing the care of higher-risk patients.

Resumen: Alrededor de 1,35 millones de personas a nivel mundial mueren anualmente por incidentes de tráfico y se estima que 50 millones sufren lesiones graves. Este panorama es particularmente dramático en la Región Andina donde el número de muertes por accidentes de tránsito asciende a 127 muertes por millón de habitantes. Recientemente, el despliegue de Sistemas Inteligentes de Transporte (SITs) en países desarrollados ha ayudado a reducir la mortalidad por accidentes de tránsito. Una parte integral de un SIT es la detección automática de incidentes de tráfico a partir de datos de video y sensores. Sin embargo, la escasez de conjuntos de datos, especialmente de casos positivos de incidentes de tráfico, obstaculizan el desarrollo de aplicaciones de inteligencia artificial para el dominio de la investigación del tráfico. En este contexto, presentamos la siguiente pregunta de investigación: ¿Es posible detectar colisiones de automóviles mediante aprendizaje automático supervisado, basado en la velocidad estimada de los autos a partir de datos de video? Como resultado presentamos VARVO, un algoritmo para la detección de incidentes de tráfico que no depende de sensores para la detección de la velocidad de los automóviles, el cual realiza una clasificación supervisada usando la detección de objetos basada en red convolucional y seguimiento bidireccional. También se describe cómo los modelos implementados en VARVO pueden mejorar su precisión de clasificación aplicando un algoritmo de sobremuestreo para clases desequilibradas. Creemos que el despliegue de VARVO podría vincularse a cámaras de video estáticas de tráfico y ser parte de los SITs en la Región Andina.

Abstract: Around 1,35 million people worldwide die each year because of traffic incidents, and it is estimated that another 50 million suffers serious injuries. This picture is particularly dramatic in the Andean Region where the death toll due to traffic accidents is as high as 127 deaths per million inhabitants. In recent years the deployment of the so-called Intelligent Transport Systems (ITS) across several developed countries has helped to reduce the number of deaths due to traffic accidents. An integral part of an ITS is the automatic detection of traffic incidents from video and sensor data. However, the scarcity of curated datasets, especially those that contained a reasonable number of positive instances of traffic incidents is hampering the development of artificial intelligence applications for the domain of traffic research. Given this scenario, we pursued answering the following research question: is it possible to detect car crashes through supervised machine learning based on the estimated speeds of cars from video only-data? Here we present VARVO, a novel algorithm for the detection of traffic incidents that does not rely on sensors for cars speed detection. VARVO performs a supervised classification task based on the sequential use of convolutional network-based object detection and bi-directional tracking. We also describe how the models implemented in VARVO improved their classification accuracy by applying an oversampling algorithm to deal with class imbalance. We believe that the deployment of VARVO could be linked to static traffic video cameras and could be part of the Intelligent Transport Systems foundations in other Andean countries.

Resumen El comportamiento de los componentes de las plantas proteicas es de vital importancia para los animales que los consumen en su dieta. La presente investigación tiene como objetivo evaluar algoritmos de regresión para determinar la conducta de las expresiones que mejor se adaptan a los procedimientos de un laboratorio tradicional y estimar los componentes químicos de plantas proteicas, en este sentido, se ha utilizado la biblioteca MULAN de java, que contiene algoritmos de aprendizaje automático capaces de adaptarse a disímiles problemas. Para ello, se crearon tres conjuntos de datos para cada especie estudiada en este trabajo; cada uno de estos incluye los elementos principales a ser evaluados en cada experimento, que están delimitados por: Metabolitos secundarios, componentes de la pared celular y digestibilidad para los ficheros de entrenamiento uno, dos y tres, respectivamente. Posteriormente, fueron evaluados por medio del aprendizaje supervisado y una validación cruzada de cada uno para determinar el mejor ajuste por aRMSE (Error cuadrático medio de la raíz). Los resultados del aprendizaje fueron comparados con experimentos anteriores, donde se tenía una variante de aprendizaje que contenía en un solo dataset todos los componentes a evaluar en una sola predicción. El resultado de la comparación muestra que los algoritmos vagos basados en instancias tienen un mejor comportamiento en el aprendizaje que los otros evaluados.

Abstract The behavior of components of protein plant is of vital importance for animals that consume them in their diet. The objective of this research is to evaluate regression algorithms, to determine the behavior of the expressions that best adapt to the procedures of a traditional laboratory and to estimate the chemical components of protein plants, in this sense the MULAN library of java has been used, that contain automatic learning algorithms capable of adapting to dissimilar problems. Three data set were created for each species treated in this study; each of these include the main elements to be evaluate in each experiment, these are delimitings by: secondary metabolites, cell wall components and digestibility element for training files one, two and three, respectively; subsequently, they were evaluated through learning supervised and cross-validation of each to determine the best fit by aRMSE (Average Root Mean Square Error). The learning results were compare with previous experiments, where there was a learning variant that contained in a single dataset all the components to be evaluates in a single prediction. The result of the comparison shows that the lazy algorithms based on instances have a better learning behavior than the others evaluate.

SUMMARY: The study purposed to examine the morphometry and morphology of crista galli in cone beam computed tomography (CBCT) and apply a new analysis, supervised Machine Learning techniques to find the answers to research questions “Can sex be determined with crista galli morphometric measurements?” or “How effective are the crista galli morphometric measurements in determining sex?”. Crista galli dimensions including anteroposterior, superoinferior, and laterolateral were measured and carried out on 200 healthy adult subjects (98 females; 102 males) aged between 18-79 years. Also, crista galli was classified with two methods called morphological types and Keros classification. In this study, the Chi-square test, Student's t-test, and Oneway ANOVA were performed. Additionally, Machine Learning techniques were applied. The means of the CGH, CGW, and CGL were found as 14.96 mm; 3.96 mm, and 12.76 mm in males, respectively. The same values were as 13.54 mm; 3.51 mm and 11.59±1.61 mm in females, respectively. The CG morphometric measurements of males were higher than those of females. There was a significant difference between sexes in terms of morphological classification type. Also, when the sex assignment of JRip was analyzed, out of 102 male instances 62 of them were correctly predicted, and for 98 female instances, 70 of them were correctly predicted according to their CG measurements. The JRip found the following classification rule for the given dataset: “if CGH<=14.4 then sex is female, otherwise sex is male”. The accuracy of this rule is not high, but it gives an idea about the relationship between CG measurements and sex. Although the issue that CG morphometric measurements can be used in sex determination is still controversial, it was concluded in the analysis that CG morphometric measurements can be used in sex determination. Also, Machine Learning Techniques give an idea about the relationship between CG measurements and sex.

RESUMEN: En el estudio se propuso examinar la morfometría y la morfología de la crista galli del hueso etmoides usando tomografía computarizada de haz cónico (CBCT) y aplicar un nuevo análisis, técnicas de aprendizaje automático supervisado para encontrar las respuestas a las preguntas de investigación "¿Se puede determinar el sexo con mediciones morfométricas de la crista galli?" o “¿Qué tan efectivas son las medidas morfométricas de la crista galli para determinar el sexo?”. Las dimensiones de la crista galli, incluidas los diámetros anteroposterior, superoinferior y laterolateral, se midieron y realizaron en 200 sujetos adultos sanos (98 mujeres; 102 hombres) con edades comprendidas entre los 18 y los 79 años. La crista galli se clasificó con dos métodos llamados tipos morfológicos y clasificación de Keros. En este estudio, se realizaron la prueba de Chicuadrado, la prueba t de Student y ANOVA de una vía. Adicionalmente, se aplicaron técnicas de Machine Learning. Las medias de CGH, CGW y CGL se encontraron en 14,96 mm; 3,96 mm y 12,76 mm en hombres, respectivamente. Los mismos valores fueron 13,54 mm; 3,51 mm y 11,59 ± 1,61 mm en mujeres, respectivamente. Las medidas morfométricas del CG de los hombress fueron más altas que las de las mujeres. Hubo una diferencia significativa entre sexos en cuanto al tipo de clasificación morfológica. Además, cuando se analizó la asignación de sexo de JRip, de 102 instancias masculinas, 62 de ellas se predijeron correctamente, y de 98 instancias femeninas, 70 de ellas se predijeron correctamente de acuerdo con las mediciones de CG. El JRip encontró la siguiente regla de clasificación para el conjunto de datos dado: "si CGH<=14.4, por tanto el sexo es femenino, de lo contrario, el sexo es masculino". La precisión de esta regla no es alta, pero da una idea de la relación entre las medidas del CG y el sexo. Aunque la pregunta si las medidas morfométricas CG se pueden usar en la determinación del sexo sigue aún siendo controvertida. Se concluyó en el análisis que las medidas morfométricas CG se pueden usar en la determinación del sexo. Además, las técnicas de aprendizaje automático dan una idea de la relación entre las medidas de CG y el sexo.

Abstract: Analysis of surface electromyography (sEMG) signals is a common practice in biomedical applications for recognizing muscle movement, wavelet coefficients obtained from wavelet transform (WT) or wavelet packet transform (WPT) are used as features of the sEMG signal and classified by means of machine learning models. To the best of our knowledge, no study has fully exploited the resemblance wavelet coefficients have to the signal from which they were obtained. In this context, time domain feature extraction on smaller data lengths can be applied directly to approximation and detail coefficients for different decomposition levels. This can be seen as different frequency band filtered versions of the original signal. The aim of this research is to compare time domain feature extraction of wavelet coefficients obtained from WT and WPT against time domain feature extraction for different frequency bands filtered sEMG signals and determine which approach is most suitable for hand movement recognition. To this end, sEMG signals were decomposed using both the WT (level 6, ‘db4’) and WPT (level 3, ‘db4’) methodologies to compare results. The comparison criterion reflects the results of the classification of three machine learning models. Results were obtained by performing supervised multiclass classifications of 18 upper limb movements from 40 subjects, retrieved from the 2nd public database generated for the Ninapro Project. The use of a lower number of coefficients can produce similar performance results as shown when comparing WT vs WPT. In the other hand, time domain feature extraction from filtered sEMG signals using wavelet reconstruction produces slightly better performance on classification results at a higher computational cost.

Abstract: There are two main approaches to automatic text classification: content-based classification and style-based classification. With content-based text classification, the topic of a document (politics, sports, health) or fake news is detected. On the other hand, Style-based text classification is used to detect the gender or age of an author, author identification, and authorship attribution. In style-based classification, the set of words defines the author’s vocabulary, which contains several hundred words. In this work, the words are known as dimensions. Texts generate high-dimensional vectors. Multiple works have shown that a large number of dimensions decreases the performance of classifiers. To reduce dimensions there are selection and extraction techniques. This article discusses the use of extraction techniques, which create low-dimensional vectors from combinations of the high-dimensional vector. Due to the development of Deep Learning networks, the use of dimensión reduction techniques has decreased because these networks perform dimensión reduction automatically. However, in Machine Learning such techniques are still used intensively. Motivated by the above, in this paper, the Principal Component Analysis (PCA) and Latent Semantic Analysis (LSA) dimensión reduction algorithms are proposed for the identification of texts written by 14 authors of the Corpus PAN 2012. The texts were divided into sequences of 10, 20, and 30 words called sentences. Likewise, blocks of texts made up of 100 sentences were created. The supervised classification was performed with the Nearest Neighbors (KNN), Support Vector Machines (SVM) and Logistic Regression (LR) algorithms using the accuracy metric. The results showed that the reduction of dimensions with PCA and the LR and SVM classifiers achieved better results than other similar works of the state of the art using the same corpus.

RESUMEN El aprendizaje conjunto es un término que se utiliza para referirse a los métodos que combinan varios algoritmos para tomar una decisión, generalmente en tareas de aprendizaje automático supervisado. En los últimos años la regresión logística, Support Vector Machine, la red neuronal y otros algoritmos de aprendizaje de máquinas se han utilizado en el diagnostico inteligente de fallas en maquinaria rotativa, sin embargo, cada algoritmo tiene sus ventajas y desventajas y pueden funcionar de distinta manera para un determinado conjunto de datos. En el presente artículo científico se proponen tres modelos de detección de fallas en cajas de engranajes bajo el enfoque de aprendizaje conjunto utilizando las técnicas de clasificación por votación mayoritaria, votación suave y apilamiento a través de las cuales se combinan las predicciones de varios estimadores a fin de mejorar la generalización y la robustez sobre un solo estimador. Para el desarrollo de los modelos especificados se utiliza un conjunto de datos de señales de falla de cajas de engranajes y cuatro técnicas de clasificación las cuales son: regresión logística, support vector machine, XGBoost y random forest. Los resultados finales de precisión de los modelos de votación mayoritaria y votación suave son 99.86% y 99.82% respectivamente, mientras que la precisión del modelo bajo el enfoque de apilamiento es de 99.85%. En comparación con los resultados de los clasificadores individuales, se demuestra que los modelos de aprendizaje conjunto compensan los errores cometidos por los clasificadores individuales y mejoran efectivamente la precisión de clasificación.

ABSTRACT Joint learning is a term used to refer to methods that combine several algorithms to make a decision, usually in supervised machine learning tasks. In recent years logistic regression, Support Vector Machine, neural network and other machine learning algorithms have been used in intelligent fault diagnosis in rotating machinery, however, each algorithm has its advantages and disadvantages and may perform differently for a given data set. In this scientific paper, three gearbox fault detection models are proposed under the ensemble learning approach using the majority voting, soft voting and stacking classification techniques through which the predictions of several estimators are combined in order to improve generalization and robustness over a single estimator. A data set of gearbox failure signals and four classification techniques are used to develop the specified models: logistic regression, support vector machine, XGBoost and random forest. The final accuracy results of the majority voting and soft voting models are 99.86% and 99.82% respectively, while the accuracy of the model under the stacking approach is 99.85%. Compared with the results of the individual classifiers, it is shown that the ensemble learning models compensate for the errors made by the individual classifiers and effectively improve the classification accuracy.

Bot detection is increasingly relevant considering that automated accounts play a disproportionate role in spreading disinformation, controlling social interactions, influencing social media algorithms and manufacturing public opinion online for different purposes. Definition, description and detection of automated manipulation techniques have proved a challenge as technology quickly advances in reach and sophistication. Considering the high contextual character of social science research, the employment of off-the-shelf detection tools raises questions regarding the applicability of machine learning systems in different cases, times and places. Thus, our purpose is to discuss the role of computational methods focusing on understanding the limitations and potential of machine learning systems to identify bots on social media platforms. To address it, we analyze the performance of Botometer, a widely adopted detection tool, in a specific domain (Amazon Forest Fires) and language (Portuguese) and propose a supervised machine learning classifier, called Gotcha, based on Botometer's framework and trained for this specific dataset. We also question how our classifier behaves and evolves over time and perform tests to evaluate the generalization capabilities of the retrained model. Our results demonstrated that supervised methods do not perform well with datasets that present features on which the system was not directly trained, such as language and topic. Hence, our study shows that a successful computational model does not always guarantee reliable results, applicable to a specific real case. Our findings indicate the need for social scientists to confirm the reliability of different tools created and tested only through the prism of computational studies before applying them to empirical social science research.

Abstract In order to defend the importance of teacher training from critical and transformative perspectives, we report in this article a research that aimed to understand the instruments that enhance such training. With a qualitative approach and a case study methodology, we analyzed how and which senses were constructed by a Physics undergraduate student during his teaching internship. We identified the instruments that influenced the student in the attribution of senses that brought him closer to being a transforming intellectual. The research was carried out at a public university in the city of Curitiba, Brazil, in the years 2019 and 2020. The empirical procedures used were: participant observations, researchers' notes, analysis of documents produced by the student during his internship, questionnaires, and semi-structured interviews. By analyzing the mobilized senses, we observed the importance of the following instruments: different ways of systematizing the intern's experiences; carrying out a collective, dialogic and autonomous teaching practice; articulation between teaching and research in teaching practices; and valuing the supervising teacher in the internship guidance process. We concluded that teaching linked to research, with strong care for the relationships between subjects in the internship, intensifies knowledge about the teaching and learning process. With these instruments highlighted, it becomes possible to rethink the organization of internship activities, seeking to develop actions that are intentionally consistent with the teacher training proposal with potentially critical and transforming senses.

Resumen Para defender la importancia de la formación docente desde perspectivas críticas y transformadoras, reportamos en este artículo una investigación que tuvo como objetivo comprender los instrumentos que potencian esa formación. Con un enfoque cualitativo y una metodología de estudio de casos, analizamos cómo y qué sentidos fueron construidos por un estudiante de grado en Física, durante su pasantía docente. Identificamos los instrumentos que influyeron en el estudiante en la atribución de sentidos que lo acercaron a un intelectual transformador. La investigación se llevó a cabo en una universidad pública de la ciudad de Curitiba, Brasil, en los años 2019 y 2020. Los procedimientos empíricos utilizados fueron: observaciones participantes, notas de las investigadoras, análisis de documentos producidos por el estudiante durante su pasantía, cuestionarios y entrevistas semiestructuradas. Al analizar los sentidos elaborados, se observó la importancia de los siguientes instrumentos: diferentes formas de sistematizar las experiencias del pasante; realizar una práctica docente colectiva, dialógica y autónoma; articulación entre docencia e investigación en la actividad docente; y la valoración del profesor supervisor en el proceso de orientación de pasantía. Concluimos que la enseñanza vinculada a la investigación, con un fuerte cuidado de las relaciones entre los sujetos en la pasantía, intensifica los conocimientos sobre el proceso de enseñanza y aprendizaje. Con estos instrumentos destacados, es posible repensar la organización de las actividades de pasantía, buscando desarrollar acciones intencionalmente coherentes con la propuesta de formación docente con sentidos potencialmente críticos y transformadores.

Resumo No sentido de defender a importância da formação de docentes em perspectivas críticas e transformadoras, relatamos neste artigo uma pesquisa que objetivou compreender os instrumentos que potencializam essa formação. Com uma abordagem qualitativa e uma a metodologia de estudo de caso, analisamos como e quais sentidos foram construídos por um estudante da licenciatura em Física, durante a realização de seu estágio de formação docente. Identificamos os instrumentos que influenciaram o discente na atribuição de sentidos que o aproximavam de um intelectual transformador. A pesquisa foi desenvolvida em uma universidade pública, na cidade de Curitiba, no Brasil, nos anos de 2019 e 2020. Os procedimentos empíricos utilizados foram: observações participantes, anotações das pesquisadoras, análise de documentos produzidos pelo estudante durante seu estágio, questionários e entrevistas semiestruturadas. Analisando os sentidos mobilizados, observou-se a importância dos seguintes instrumentos: diferentes formas de sistematização das vivências do estagiário; realização de uma prática de docência coletiva, dialógica e com autonomia; articulação entre ensino e pesquisa na atuação docente; e valorização do professor supervisor no processo de orientação do estágio. Concluímos que o ensino atrelado à pesquisa, com um acentuado cuidado para as relações entre os sujeitos no estágio, intensifica os conhecimentos sobre o processo de ensino e aprendizagem. Com esses instrumentos destacados, torna-se possível repensar a organização das atividades de estágio, buscando desenvolver ações intencionalmente condizentes com a proposta de formação de docentes com sentidos potencialmente críticos e transformadores.

Resumen Contexto: En el sector educativo se han identificado variables que inciden considerablemente en el rendimiento academico de los estudiantes. En la ultima decada se han llevado a cabo investigaciones desde diversos campos como la psicologia, la estadistica y el analisis de datos con el fin de predecir el rendimiento academico. Metodo: La analitica de datos, especialmente a traves de herramientas de Machine Learning, permite predecir el rendimiento academico utilizando algoritmos de aprendizaje supervisado basados en variables academicas, demograficas y sociodemograficas. En este trabajo se seleccionan las variables mas influyentes en el transcurso de la vida academica de los estudiantes mediante metodos de filtro, embebidos, y de ensamble, asi como las caracteristicas mas importantes semestre a semestre utilizando algoritmos de Machine Learning (arbol de decision, KNN, SVC, Naive Bayes, LDA), implementados en el lenguaje Python. Resultados: Los resultados del estudio muestran que el KNN es el modelo que mejor predice el rendimiento academico para cada uno de los semestres, seguido de los arboles de decision, con valores de precision que oscilan alrededor del 80 y 78,5% en algunos semestres. Conclusiones: Con respecto a las variables, no se puede decir que el promedio academico semestral de un estudiante influya necesariamente en la prediccion del rendimiento academico del siguiente semestre. El analisis de estos resultados indica que la prediccion del rendimiento academico utilizando herramientas de Machine Learning es un enfoque promisorio que puede ayudar a mejorar la vida academica de los estudiantes y permitir a las instituciones y a los docentes adoptar acciones que ayuden al proceso de ensenanza-aprendizaje.

Abstract Context: In the education sector, variables have been identified which considerably affect students’ academic performance. In the last decade, research has been carried out from various fields such as psychology, statistics, and data analytics in order to predict academic performance. Method: Data analytics, especially through Machine Learning tools, allows predicting academic performance using supervised learning algorithms based on academic, demographic, and sociodemographic variables. In this work, the most influential variables in the course of students’ academic life are selected through wrapping, embedded, filter, and assembly methods, as well as the most important characteristics semester by semester using Machine Learning algorithms (Decision Trees, KNN, SVC, Naive Bayes, LDA), which were implemented using the Python language. Results: The results of the study show that the KNN is the model that best predicts academic performance for each of the semesters, followed by Decision Trees, with precision values that oscillate around 80 and 78,5% in some semesters. Conclusions: Regarding the variables, it cannot be said that a student’s per-semester academic average necessarily influences the prediction of academic performance for the next semester. The analysis of these results indicates that the prediction of academic performance using Machine Learning tools is a promising approach that can help improve students’ academic life allow institutions and teachers to take actions that contribute to the teaching-learning process.