RISTI - Revista Ibérica de Sistemas e Tecnologias de Informação

Fifteen years of RISTI - Revista Ibérica de Sistemas e Tecnologias de Informação: fifty regular issues and sixty special issues

A systematic review of dengue classification models using machine learning.

El dengue es una enfermedad arboviral que anualmente reporta un gran número de infectados en la costa norte y la selva peruana. Según las estadísticas, está aumentando cada año. Este artículo tiene como objetivo desarrollar una revisión sistemática de la literatura científica sobre las variables de estudio y los métodos de aprendizaje automático utilizados actualmente para detectar la infección por dengue. La metodología utilizada fue PRISMA, mapeando inicialmente la literatura de 274 artículos científicos, quedando seleccionados 33 artículos para la revisión sistemática. Los resultados obtenidos son que los algoritmos de aprendizaje automático más utilizados son las redes neuronales (NN) y support vector machine (SVM). Asimismo, se ha encontrado que los científicos tienden a realizar investigaciones con variables climáticas o demográficas para obtener mejores resultados. Se concluye que los métodos de aprendizaje automático que más se han utilizado son las redes neuronales de diferentes tipos: convolucional, recurrente, profunda y multicapa, y para la predicción de brotes de dengue predominaron los métodos de series de tiempo con LSTM y ARIMA, también se estableció que la tendencia es hacia la inclusión de variables climáticas y demográficas en los modelos de predicción.

Abstract Dengue is an arboviral disease that annually reports a large number of infected on the north coast and the Peruvian jungle. According to statistics, it is increasing yearly. This article aims to develop a systematic review of the scientific literature on the study variables and the machine learning methods currently used for detecting dengue infection. The methodology used was PRISMA, initially mapping the literature of 274 scientific articles, leaving 33 articles selected for the systematic review. The results obtained are that the most used machine learning algorithms are neural networks (NN) and support vector machine (SVM). Likewise, it has been found that scientists tend to carry out research with climatic or demographic variables to obtain better results. It is concluded that the machine learning methods that have been used the most are neural networks of different types: convolutional, recurrent, deep, and multilayer, and for the prediction of dengue outbreaks the time series methods with LSTM and ARIMA were the predominant ones, it was also established that the trend is towards the inclusion of climatic and demographic variables in the prediction models.

Influence of e-government related indicators on Portugal's Municipal Transparency Index (MTI)

Resumo Em que medida o Índice de Transparência Municipal (ITM) de Portugal é afetado pelos indicadores relacionados ao efeito esperado do governo eletrônico na transparência? Há inter-relacionamento entre eles? O presente artigo se propõe a responder a estas questões, utilizando-se do referencial teórico relacionado à transparência e ao governo eletrônico, que trata da importância da transparência, dos efeitos do governo eletrônico na transparência, do conteúdo dos sites governamentais e de como o governo eletrônico pode ajudar a melhorar a transparência. Por meio de procedimentos estatísticos, tratou-se de identificar a capacidade dos 15 indicadores selecionados para explicar o resultado do ITM, além de identificar similaridades entre os municípios portugueses e entre os indicadores. Os resultados mostraram que nos indicadores selecionados há poder de explicação considerável do ITM. Foi possível ainda identificar, dentre os 15, aqueles indicadores com maior poder de explicação do ITM bem como as similaridades entre os indicadores.

Abstract To what extent is Portugal's Municipal Transparency Index (MTI) affected by the indicators related to the expected effect of e-government on transparency? Is there an interrelationship between them? This paper proposes to answer these questions, using the theoretical framework related to transparency and e-government, which deals with the importance of transparency, the effects of e-government on transparency, the content of government websites and how e-government can help improve transparency. Through statistical procedures, we tried to identify the ability of the 15 selected indicators to explain the MTI score, and to identify similarities among the Portuguese municipalities and among the indicators. The results showed that in the selected indicators there is considerable explanatory power of the MTI. It was also possible to identify, among the 15, those indicators with greater power to explain the MTI as well as the similarities among the indicators.

Effects of using augmented reality on students learning: a systematic review

Resumo Este artigo apresenta os resultados de uma revisão sistemática da literatura sobre os efeitos da utilização da realidade aumentada (RA) na aprendizagem dos estudantes de ciências (Ciências Naturais e Biologia) do 3.ºciclo do ensino básico e do ensino secundário. Foi utilizada a metodologia PRISMA, tendo sido selecionados 10 artigos de revistas indexadas, num período de tempo entre 2010 e 2022. As bases de dados consultadas foram Scopus, ScienceDirect, Springer Link, EBSCO e Web of Science. A maioria dos artigos selecionados (n=8), após a aplicação dos critérios de inclusão e exclusão, utilizou uma metodologia experimental. Os resultados indicam que a RA teve efeitos positivos na aprendizagem e motivação dos estudantes, bem como noutras variáveis como as competências visuoespaciais e o envolvimento dos estudantes nas tarefas de aprendizagem.

Abstract This article presents the results of a systematic review of the literature regarding the effects of using augmented reality (AR) on the learning of science students (Natural Sciences and Biology) in lower and high secondary education. The PRISMA methodology was used. Ten articles were selected from indexed journals, in a period of time between 2010 and 2022. The databases consulted were Scopus, ScienceDirect, Springer Link, EBSCO, and Web of Science databases. Most articles selected (n=8), after applying the inclusion and exclusion criteria, used an experimental methodology. The results indicate that AR had positive effects on students' learning and motivation, as well as on other variables such as visuospatial skills and student involvement in tasks.

Automatic Grading Systems for Programming Assignments: Review and Critique

Resumen La corrección de las tareas de programación de computadoras es una actividad crítica y de alto consumo de tiempo para los académicos. Como respuesta a este problema, se han propuesto diversos sistemas de retroalimentación y calificación automática. Este artículo presenta una revisión de sistemas y enfoques de corrección y calificación automatizada de tareas de programación con el objetivo de ofrecer un mapa general de las funciones de estos sistemas y sus desafíos. Los resultados muestran que existe una importante diferencia entre la industria y la academia, así como una carencia en marcos de evaluación para estas herramientas.

Abstract: Reviewing programming assignments has become a time-consuming and critical activity for academics. Various feedback and automatic grading systems have been proposed in response to this problem. This article presents a review of various systems and automated feedback approaches for grading programming assignments to provide an overview of the functions of these systems and their challenges. The results show a difference between the tools proposed by the academy and those proposed by the industry, as well as a lack of evaluation frameworks.

Trend Model for complex series: case study hotel rates

Resumen El análisis riguroso de datos en turismo es cada vez más común a raíz de las afectaciones sufridas por el covid-19, lo que ha contribuido a identificar trayectorias a partir de datos históricos. En este artículo, se propone una combinación de la teoría de la complejidad y la suavización exponencial para conocer los patrones que ayuden a estimar la estacionalidad y tendencia en las tarifas hoteleras. Para ello, se fragmenta la serie en particiones para explorar la similitud y se aplican cinco métodos de trazado. Los resultados muestran que el gráfico de longitud R, es el único que despliega anti-persistencia, con una saturación en el tiempo n=8, este dato se usa como ciclo de longitud del modelo aditivo de Holt-Winters cuyos parámetros (a=0.36, b=0.78, y=0.67), minimizan el error presentado. Se concluye que la combinación de métodos de análisis numérico permite el ajuste estacional en los datos tarifarios.

Abstract In The rigorous analysis of data in tourism is increasingly common as a result of the effects suffered by covid-19, which has contributed to identifying trajectories based on historical data. In this article, a combination of complexity theory and exponential smoothing is proposed to understand the patterns that help to estimate the seasonality and trend in hotel rates. To do this, the series is fragmented into partitions to explore the similarity and five plotting methods are applied. The results show that the graph of the length R, is the only one that displays anti-persistence, with a saturation in time n=8, this data is used as the length cycle of the additive Holt-Winters model whose parameters (a= 0.36, b=0.78, y=0.67), minimize the error presented. It is concluded that the combination of numerical analysis methods allows the seasonal adjustment in the tariff data.

Divergence in e-government maturity levels in Mexico

Resumen La presente investigación busca de responder a dos ideas centrales, primero, demostrar que existen sitios web de e-gobierno en un mismo nivel de madurez con extremas divergencias; segundo, proponer las categorías que se desprenden de un mismo nivel de madurez de e-gobierno con la finalidad de conocer de una forma precisa el estatus de un sitio de e-gobierno a nivel municipal. El método utilizado fue el descriptivo, analizando los 2470 municipios que conforman a las 32 entidades de México. Los hallazgos demuestran una divergencia importante entre los sitios web ubicados en un mismo nivel de los modelos madurez, principalmente aquellos que se encuentran en el nivel interactivo, lo cual, sugiere que existen tipos de sitios web interactivos, unos con funciones asincrónicas y otros con sincrónicas. Concluyendo que existe un nivel de interacción básico y otro avanzado. Este análisis contribuye a la investigación ofreciendo avances teóricos relacionados con las etapas de e-gobierno.

Abstract The present research seeks to respond to two central ideas: first, to demonstrate that there are e-government websites at the same level of maturity with extreme divergences; second, to propose the categories that arise from the same e-government level maturity to precisely know the status of an e-government site at the municipal level. The method used was descriptive, analyzing the 2,470 municipalities that make up the 32 Mexican states. The findings show a significant divergence between websites located at the same level of the maturity models, mainly those at the interactive level, which suggests that there are types of interactive websites, some with asynchronous functions and others with synchronous ones. Concluding that there is a basic and an advanced level of interaction. This analysis contributes to the research by offering theoretical advances related to the stages of e-government.

The effect of a telehealth psychotherapy program on the mental health of youth at high psychosocial risk

Resumen Este estudio experimental tuvo como objetivo determinar el efecto de la psicoterapia de telesalud en la salud mental de jóvenes en riesgo. La salud mental se evaluó utilizando el Cuestionario de Salud General de Goldberg. La intervención consistió en talleres sincrónicos realizados por personal experto en enfermería, mientras que el grupo de control no recibió intervención alguna. Se utilizó la prueba de McNemar para evaluar las diferencias entre las pruebas previas y posteriores en cada grupo. La muestra comprendió a 60 individuos jóvenes. Antes de la intervención, se encontraron síntomas psicopatológicos en el 67% y 93% de los grupos de control y experimental, respectivamente. Después de la intervención, estas frecuencias disminuyeron al 73% y 27%, respectivamente. La diferencia no fue significativa en el grupo de control (p = 0.316), pero sí fue significativa en el grupo experimental (p = 0.000025). En conclusión, la psicoterapia de telesalud dirigida por profesionales de enfermería mejora significativamente la salud mental de jóvenes en riesgo.

Abstract This experimental study aimed to determine the effect of telehealth psychotherapy on the mental health of at-risk youth. Mental health was assessed using the Goldberg General Health Questionnaire. The intervention consisted of synchronous workshops conducted by expert nursing staff, while the control group received no intervention. The McNemar test was used to evaluate the differences between pre- and post-tests within each group. The sample comprised 60 young individuals. Prior to the intervention, psychopathological symptoms were found in 67% and 93% of the control and experimental groups, respectively. After the intervention, these frequencies decreased to 73% and 27%, respectively. The difference was not significant in the control group (p = 0.316), but it was significant in the experimental group (p = 0.000025). In conclusion, telehealth psychotherapy led by nursing professionals significantly improves the mental health of at-risk youth.

Conversational systems: cultural aspects influence the UX evaluation of a conceptual voice assistant prototype

Resumo Este artigo analisa a experiência de 100 indivíduos (50 brasileiros e 50 portugueses) no contexto de um sistema conversacional com assistente de voz automatizado. A avaliação da UX foi realizada em diferentes etapas: cálculo da estatística descritiva, análise das frases ditas pelos utilizadores e aplicação da escala SUS. Esta abordagem metodológica permitiu verificar que o protótipo conceptual simula um produto com grande potencial para facilitar o dia a dia dos profissionais que atuam em salões de estética e beleza. No entanto, há funcionalidades que devem ser redesenhadas, como a escolha dos nomes dos botões e a acessibilidade da ferramenta. Outra contribuição do estudo é a confirmação de que há opiniões divergentes entre as populações da amostra, brasileira e portuguesa, mostrando que a solução teve melhor aceitação entre os portugueses. Esta última questão é um dos aspetos mais relevantes dos achados deste trabalho, pois pode comprometer o sucesso de tal assistente.

Abstract This paper analyzes the experience of 100 individuals (50 Brazilians and 50 Portuguese) in the context of a conversational system with an automated voice assistant. The UX evaluation consisted of calculating descriptive statistics, analyzing the phrases spoken by users during the tests and applying the SUS scale. This methodological approach made it possible to verify that the conceptual prototype simulates a product with great potential for facilitating the daily lives of professionals working in beauty salons. However, there are features that need to be redesigned, such as the choice of button names and the accessibility of the tool. Another contribution is the confirmation that there are divergent opinions between the Brazilian and Portuguese sample populations, showing that the solution was better accepted by the Portuguese. This last issue is one of the most relevant aspects of this study's findings, as it could jeopardize the success of such an assistant.