Em disciplinas de programação, um único exercício pode ser resolvido de diferentes formas. Compreender as estratégias adotadas pelos estudantes para solucionar problemas é pedagogicamente relevante, pois permite avaliar, por exemplo, se eles estão assimilando os conteúdos abordados em sala e aplicando-os corretamente em seus códigos. Para ajudar os professores nesse contexto, o presente artigo investiga o uso de LLMs para rotular códigos de alunos de acordo com a estratégia que eles usaram para solucionar os exercícios de programação. Para tanto, foi criada uma base de dados com códigos rotulados e posteriormente realizados experimentos com diferentes LLMs. Os resultados preliminares indicam que, com a formulação de prompts adequados, os LLMs têm potencial para desempenhar a tarefa de rotulagem automática de códigos.

Para melhorar a qualidade do código dos estudantes, pesquisadores buscam alternativas de fornecer feedback sobre o código não somente quanto à corretude, mas também sobre a sua qualidade. Entretanto, percebe-se que a realidade de salas de aula com muitos alunos não permite um feedback individual e completo. Neste contexto, ferramentas como Analisadores Estáticos (AEs) podem ser utilizadas para realizar a análise do código sem executá-lo, fornecendo relatórios sobre sua qualidade como, por exemplo, problemas de estilo. A partir deste entendimento, o presente trabalho apresenta uma ferramenta para avaliação da qualidade de código em relação ao uso de convenções de estilo de linguagens de programação. Para tanto, criamos métricas para resumir a quantificação da qualidade do código a partir dos resultados dos AEs. Por fim, criamos um ambiente para simplificar a análise científica de datasets de código de estudantes. O trabalho apresenta uma descrição detalhada do ambiente, além de exemplos ilustrativos do seu funcionamento.

Juízes online, amplamente usados em disciplinas introdutórias de programação, costumam focar na conformidade com casos de teste, pouco contribuindo para identificar compreensões imperfeitas (misconceptions) do aprendiz sobre o assunto. Neste trabalho, levantou-se a evolução dos misconceptions exibidos pelos alunos ao longo de um período letivo em uma disciplina desse tipo, a partir dos códigos submetidos a um juiz online. Os resultados indicam que misconceptions relacionados a estruturas de decisão são os mais persistentes ao longo do tempo. Além disso, verificou-se que alguns misconceptions frequentemente surgem em conjunto, como os ligados a atribuições e estruturas redundantes, aplicadas desnecessariamente na codificação, enquanto outras categorias aparecem pontualmente. A contribuição desta análise não se limita a catalogar os misconceptions dos alunos, possibilitando que pesquisas futuras abordem estratégias de mitigar os mais frequentes e os mais recorrentes.

The mobile app market has increased substantially in the past decades, and the myriad options in the app stores have made users less tolerant of low-quality apps. In this competitive scenario, User eXperience (UX) has emerged as an essential factor in standing out from competitors. By understanding what factors affect UX, practitioners could focus on factors that lead to positive UX while mitigating those that affect UX negatively. In this context, app store reviews emerged as a valuable resource for investigating these influential factors. However, analyzing millions of reviews can be costly and time-consuming. This article introduces UX-MAPPER, a tool designed to analyze app store reviews and assist practitioners in pinpointing factors that impact UX. We applied the Design Science Research method to develop UX-MAPPER iteratively and rooted in a robust theoretical background. We performed exploratory studies to investigate the problem, a systematic mapping study to identify UX-affecting factors, and an empirical study to ascertain practitioners’ relevance and acceptance of UX-MAPPER. In general, the participants recognized the relevance and utility of UX-MAPPER in enhancing the quality of existing apps and exploring reviews of competing apps to identify user preferences, requests, and critiques regarding functionalities and features. However, the output quality requires refinement to better convey the benefits of the results, especially for practitioners with prior experience with automated approaches. From the participants’ feedback, we defined a set of suggestions to extract more useful features, which can contribute to future studies involving user review analysis. Based on the results of this research, we present the contributions to the area of HCI and possible developments for future research.

Blended learning can be considered an approach that combines face-to-face and online periods in education through the integration of some technological resources, such as media centers. The present research sought to investigate the panorama of scientific production on blended learning and identify evidence of how media centers are used to support this type of teaching. This is a bibliometric study carried out based on the principles of the Systematic Literature Review (SLR). Although the results indicate an increase in production, a high rate of authors with only one published work was identified, which can be interpreted as an expanding topic. The United States stood out as the country that contributed the most with research and the areas of Computer Science and Education among the fields of research with the highest number of articles.

As contribuições de Game Learning Analytics (GLA) aliadas a testes heurísticos podem fornecer insights aos educadores sobre possíveis evidências de aprendizagem. Entretanto, quando se tenta incluir técnicas de captura de dados após implementação do jogo, podem surgir desafios. Na tentativa de mensurar tais desafios, este trabalho apresenta um estudo de caso que explora os desafios encontrados na implementação da ferramenta GLBoard em um jogo educacional parcialmente desenvolvido. Resultados apontam que, apesar da participação de dois especialistas, a implementação apresentou desafios: abstração do jogo, mecânica complexa, adaptação da estrutura e navegação pelo código.

Due to their versatility, concept maps are used in various educational settings and serve as tools that enable educators to comprehend students' knowledge construction. An essential component for analyzing a concept map is its structure, which can be categorized into three distinct types: spoke, network, and chain. Understanding the predominant structure in a map offers insights into the student's depth of comprehension of the subject. Therefore, this study examined 317 distinct concept map structures, classifying them into one of the three types, and used statistical and descriptive information from the maps to train multiclass classification models. As a result, we achieved an 86% accuracy in classification using a Decision Tree. This promising outcome can be employed in concept map assessment systems to provide real-time feedback to the student.

Game designers and researchers have sought to create gameful environments that consider user preferences to increase engagement and motivation. In this sense, it is essential to identify the most suitable game elements for users' profiles. Designers and researchers must choose strategies to classify users into predefined profiles and select the most appropriate game elements for each user. This activity may challenge designers, learning designers, and researchers since they must base their choice on personal aspects that require a deep understanding. Therefore, this article aims to assist game designers, learning designers, and researchers in selecting user classification strategies to customize and personalize game-based and gamified learning environments. By conducting systematic literature mapping, we consolidate the most common strategies and explore their applications in games and gamification. Our analysis, based on 25 publications, reveals that we can classify the strategies according to user interaction, user personality, learning style, and motivation for learning. Strategies based on user interactions emerge as the most popular, while questionnaires and log data systems are commonly used instruments for identifying user profiles. The findings of this SLM offer valuable knowledge for game designers and researchers to define the criteria that will be used to evaluate the effect of games and gamified environments in educational contexts.

Gamification applied to learning environments is widely accepted as positively impacting students' psychological and cognitive aspects, such as motivation and learning performance. According to the literature on the subject, gamification tends to promote more positive effects on students than negative ones. Meanwhile, the literature lacks a deeper understanding of how education professionals perceive gamification in learning environments and their concerns about implicit issues and ethical issues. Prior research has not examined the relationship between gamification in education, its ethical concerns, and barriers. As a result, we expanded a previous study to identify and delve deep into potential barriers and ethical concerns pertaining to gamification from the perspective of Brazilian teachers. A survey was designed and answered by 61 Brazilian teachers. According to our findings, teachers are not inclined to use gamification for various reasons, such as social factors (e.g., acceptance by teachers and students) and planning and evaluation issues (e.g., lack of knowledge). Our study also found that their ethical concerns pertain to psychological effects, social issues, privacy issues, humanization, and behavioral effects. As part of the contribution of this paper, we list potential barriers and ethical concerns that designers and researchers should keep in mind when designing and implementing gamification and gamification-based personalization in learning environments.

A avaliação da carga cognitiva em jogos é essencial para alinhar o design de interação à capacidade cognitiva do jogador, assegurando imersão e facilitando a aprendizagem. No entanto, a inclusão de elementos não contributivos, especialmente em contextos complexos como problemas de computação, pode prejudicar a aprendizagem. O objetivo deste estudo é investigar como a carga cognitiva influencia o design do jogo e a perspectiva dos jogadores. Para isso, um jogo sobre Problema da Mochila é analisado, ao qual utilizou-se o NASA-TLX e o MEEGA+ para avaliar as dimensões associadas ao esforço cognitivo de estudantes de cursos de computação do 3º ao 8º período. Resultados indicam que a carga cognitiva foi considerada moderada pelos estudantes, mas que ainda é desbalanceada em algumas fases, sugerindo a presença de elementos que causam sobrecarga cognitiva.

Introduction and aimsObesity is a multifactorial condition with high health risk, associated with important chronic disorders such as diabetes, dyslipidemia, and cardiovascular dysfunction. Citrus aurantium L. (C. aurantium) is a medicinal plant, and its active component, synephrine, a β-3 adrenergic agonist, can be used for weight loss. We investigated the effects of C. aurantium and synephrine in obese adolescent mice programmed by early postnatal overfeeding.MethodsThree days after birth, male Swiss mice were divided into a small litter (SL) group (3 pups) and a normal litter (NL) group (9 pups). At 30 days old, SL and NL mice were treated with C. aurantium standardized to 6% synephrine, C. aurantium with 30% synephrine, isolated synephrine, or vehicle for 19 days.ResultsThe SL group had a higher body weight than the NL group. Heart rate and blood pressure were not elevated. The SL group had hyperleptinemia and central obesity that were normalized by C. aurantium and synephrine. In brown adipose tissue, the SL group showed a higher lipid droplet sectional area, less nuclei, a reduction in thermogenesis markers related to thermogenesis (UCP-1, PRDM16, PGC-1α and PPARg), and mitochondrial disfunction. C. aurantium and synephrine treatment normalized these parameters.ConclusionOur data indicates that the treatment with C. aurantium and synephrine could be a promising alternative for the control of some obesity dysfunction, such as improvement of brown adipose tissue dysfunction and leptinemia.

Digital games are one of the most prominent forms of entertainment in the contemporary world. For those that are inspired by such games, there's the possibility of a career path in the production of digital games. In this work, we aim to report on the pre-production stage of our first commercial game, which is "Meu jardinzinho" (Portuguese for "My little garden"). By briefly describing our ideation, design and prototyping process, we intend to discuss the results from what we learned from Scott Rogers' book, "Level Up! The guide to great video game design", through our lens as beginner game designers. Our results tell us that, even by following the steps proposed in the book, the execution is not straightforward, requiring more experience and rework than anticipated.

Ambientes de correção automática de código são cada vez mais usados no processo de ensino-aprendizagem de disciplinas de programação. Porém, um problema frequentemente enfrentado pelos professores que usam tais ambientes é determinar a dificuldade das questões cadastradas. Este trabalho tem como objetivo realizar uma análise de correlação entre métricas de complexidade de código e a dificuldade enfrentada pelos alunos, de maneira que seja possível prever automaticamente o nível de dificuldade de uma questão apenas conhecendo seu modelo de solução. Este estudo foi dividido em três etapas: i) análise da correlação de Spearman entre métricas de complexidade (extraídas da questão) e de dificuldade (extraídas da interação do aluno com a questão); ii) predição da classe de dificuldade de questões por meio de modelos de aprendizado de máquina para classificação; e iii) predição de métricas de dificuldade usando modelos de regressão. Quanto ao item i), observou-se que 96% das correlações foram fracas ou inexistentes entre métricas individuais de complexidade de código e de dificuldade, 4% de casos de correlação moderada e nenhum caso de correlação forte. Para o item ii), o maior f1-score obtido foi de 88%, considerando classificação com dois níveis de dificuldade (“fácil” e “difícil”), e f1-score máximo de 67%, considerando classificação com três níveis (“fácil”, “médio” e “difícil”). Para o item iii), o melhor resultado obtido foi um coeficiente de determinação ajustado de 63%.

A coleta de dados educacionais é essencial para a gestão eficiente de recursos e atendimento das necessidades da população, sendo amplamente utilizada por pesquisadores para compreender e melhorar a educação. Nesse contexto, este estudo visa identificar quais os fatores de risco de evasão escolar de maior impacto com base no Instrumento de Avaliação de Risco de Evasão Escolar (IAFREE). Foram realizadas análises estatísticas e classificatórias com aprendizado de máquina para compreender as variáveis com maior significância correlacional, ao final destacando-se as relações do estudante com o ambiente familiar e a distância da casa até a escola.

Introduction: Maternal obesity has been positively correlated with an increased cardiometabolic risk in the offspring throughout life, implying intergenerational transmission. However, little is known about the early-life cardiac cell modifications that imply the onset of heart diseases later in life. This study analyzed cardiac progenitor cells and cardiomyocyte differentiation on day of birth in the offspring born to obese dams. Methods: The litter size reduction model was used to induce obesity in female Swiss mice. Both maternal groups, the Small Litter Dams (SLD-F1), which were overfed during lactation, and the Normal Litter Dams (NLD-F1), control group, were mated to healthy male mice. Their first-generation offspring (SLD-F2 and NLD-F2, n = 6 by group) were euthanized on birth. Results: Mothers from SLD had increased body mass, Lee Index, fat deposits, hyperglycemia, and glucose intolerance, confirming the obese phenotype. The offspring born from SLD-F1 had also increased body mass, Lee Index, and fasting hyperglycemia. The heart of SLD-F2 showed decreased cardiac mass/body mass ratio, increased cardiac collagen deposits, a greater number of undifferentiated cardiac c-kit+ and Sca-1+ progenitor cells, and increased NKX2.5+ cardiomyoblasts compared to control. In addition, SLD-F2 demonstrated immature cardiomyocytes. Conclusions: Obese dams negatively impact their offspring, leading to altered biometric and metabolic parameters, along with an immature heart already at birth, with extracellular matrix adverse remodeling, delayed cardiac progenitor cell differentiation, and restrained cardiomyocyte maturation, which can be related to the development of cardiometabolic disease in the adulthood.

Online Judges (OJs) have gained substantial traction in programming education due to their ability to simultaneously present problem-solving challenges to students while offering instant feedback and correction. Such technologies are also essential to allow students in remote areas to access quality and equitable education. Nonetheless, OJ systems often lack sufficient amounts of annotated data (i.e., labelled data) about the topics of the problems that they aim to support, which makes choosing appropriate problems hard. Topic annotations hold significant value for instructors when selecting problems for assignments and for novice students seeking independent use of OJ systems. In this work, we propose and evaluate a pre-trained deep learning architecture and an active learning methodology to automatically annotate OJ problems in the context of introductory programming. Our results show that, when using a smaller amount of data, the methodology demonstrates performance comparable to those of the existing state-of-the-art methods for the identical task.

Game Learning Analytics (GLA) involves capturing and analyzing data from educational games, enabling the identification of evidence of learning. A fundamental step before implementing GLA techniques is data modeling, which is not trivial. Using large language models (LLMs) can help in this context, as they can generate text like humans. Therefore, considering Chat-GPT and its customizable functionality, “MyGPTs,” this work proposes creating a specialist agent to assist learning designers in data modeling and implementing GLA techniques based on the GLBoard system. Preliminary results with GLA specialists were positive, indicating the agent’s potential.

User eXperience (UX) evaluations play an essential role in the software development process. As the results from such evaluations can drive future releases, it is necessary to identify which factors can substantially change users' judgments about their experience to have more precise results and understand UX better. This article investigates how interaction sequencing, previous experience, and the number of problems could affect overall satisfaction and the two main UX dimensions: pragmatic and hedonic. We employed three different evaluation methods to evaluate a chatbot-based mobile shopping application. The results revealed that participants with previous experience with similar apps tended to give lower ratings. We also found that as inspectors identify more problems, they tend to rate the pragmatic dimension lower. Finally, we did not identify a significant influence of interaction sequencing on UX evaluation. We discuss the reasons for these results, the implications for practitioners and researchers, and research opportunities.

O ensino de computação na educação básica tem sido amplamente discutido devido à rápida evolução tecnológica. Este estudo apresenta um mapeamento sistemático que investiga práticas, ferramentas e métodos para aprimorar o ensino de programação em escolas ao redor do mundo. Os resultados do mapeamento sistemático revelam diversas abordagens adotadas em relação aos métodos e ferramentas utilizados, bem como uma convergência de medidas recomendadas para o avanço do ensino de programação. Esse mapeamento oferece insights valiosos para educadores interessados em promover o ensino de programação de forma eficaz, e a necessidade de se pensar na importância de se incluir esse tema nos currículos.