Interoperabilidade, a operação coordenada de sistemas, é um conceito complexo com definições variadas. Objetivo: Este estudo investiga a evolução da pesquisa em interoperabilidade nos últimos 50 anos. Método: A avaliação de metadados biliométrica foi utilizada como método. A análise abrange 21.431 publicações. Resultados: Os resultados indicam uma contribuição significativa da ciência da computação, engenharia e matemática. Conclusão: Os resultados sugerem uma alta dispersão das pesquisas sobre interoperabilidade por fontes de publicações e áreas de conhecimento e uma predominância de propostas de soluções pontuais em diferentes contextos onde a interoperabilidade é apenas uma característica, em oposição às pesquisas conceituais sobre interoperabilidade.

Learning Analytics (LA) is one of the world’s most influential research fields related to educational technology. Among many themes that the LA community considers, the application of Natural Language Processing (NLP) algorithms has been largely adopted to extract information from textual data generated in learning environments (e.g., student essays and short answers, online discussion and chat). NLP can shed light on the learning process and student outcomes in different contexts. Based on the importance of NLP for education, this paper conducted a systematic literature review of the application of NLP to understand how the LA community has been applying this method. Our methodology includes automatic and manual methods to extract information about authors, relevant papers, and specific data related to educational applications and algorithms used in the field. This review selected 156 papers that reveal essential aspects of the topic, such as: (i) the majority of the works focused on the analysis of online discussions and essay assessment; (ii) in general, the authors did not apply the developed models in real settings; (iii) recent papers selected start to evaluate deep learning models (e.g., BERT) more frequently; (iv) the datasets used in the experimentation are usually small and containing English text; (v) the average models performance reaches 0.54 and 0.79 of Cohen’s Kappa and Accuracy, respectively. The results of this study and its practical implications are further discussed.

Learning analytics (LA) adoption is a challenging task for higher education institutions (HEIs) since it involves different aspects of the academic environment, such as information technology infrastructure, human resource management, ethics, and pedagogical issues. Therefore, it is necessary to provide institutions with supporting instruments to deal with these challenges. Although there has been much research on factors that are associated with the adoption of LA in HEIs, there has been much less research on specific models that can be used to guide actual adoption. In this sense, we developed MMALA, a Maturity Model for Adopting Learning Analytics. It is a guide that describes the necessary practices for taking the first steps in this area and enables institutions to reach higher levels of maturity in LA use, culminating in an organized and systematic adoption. In this paper, we describe the development process of MMALA, focusing on the model evaluation, which used both the questionnaire and the expert opinion method. MMALA can also give institutions an overview of their current situation regarding LA adoption. In this sense, we present the results of the maturity evaluation of three Brazilian HEIs using MMALA.

Accurate punctuation in written text enables unambiguous communication, minimizing the risk of misunderstandings. Conversely, faulty punctuation can confuse the intended meaning, posing challenges for the author. The existing literature offers a collection of systems and algorithms to assist users in writing tasks. However, those focusing on English tend to exhibit higher accuracy. Furthermore, most models for punctuation restoration yield results without offering insight into their decision-making processes. Therefore, this study evaluated state-of-the-art punctuation restoration models specifically for Brazilian Portuguese and incorporated the principles of explainable artificial intelligence to clarify their predictions transparently. The findings indicate that the models assessed achieved an accuracy comparable to those of their English-language counterparts.

Textual production is a key activity at different levels of education. The analysis of essays encompasses several criteria, such as lexical and syntactic errors, cohesion, and coherence. Within these criteria, how the students include punctuation (i.e., final mark and comma) could influence the quality of the final production. Thus, the literature has proposed several approaches to verifying punctuation correction in students’ essays for English. However, despite the advancements in natural language processing models for other languages, there is a significant gap concerning punctuation verification. Therefore, this paper proposed a new approach based on state-of-the-art language models to develop a punctuation prediction method for Portuguese. The proposed model was applied to evaluate the textual productions of students in Brazilian public schools. Finally, the results of this study and its practical implications for educational settings are further discussed.

A área de Inteligência Artificial (IA) tem potencial para melhorar o ensino e a aprendizagem, por exemplo, por meio da análise de dados produzidos em ambientes educacionais. Além disso, também pode agravar a desigualdade, pois exige que alunos e instrutores tenham acesso à infraestrutura (smartphones ou computadores) exigida pela maioria dessas ferramentas para gerar e analisar dados. No entanto, o acesso a tal infraestrutura não é uma realidade para muitos estudantes ao redor do mundo. Para lançar luz sobre esse problema, este artigo investiga, por meio de um Estudo de Mapeamento Sistemático (MS), iniciativas que permitem uma análise de dados mais inclusiva usando IA na educação, especialmente em cenários com poucos recursos de conectividade. Identificamos que essas iniciativas são escassas e estão focadas na primeira fase da tarefa de análise de dados: a coleta de dados. Com base nos resultados do MS, propomos um conjunto de recomendações para os pesquisadores oferecerem direções para uma análise mais inclusiva de dados educacionais usando IA.

A Avaliação Automática de Redação (do inglês, Automatic Essay Scoring - AES) tem sido tema amplamente explorado na literatura. Ela permite dispensar o esforço humano aplicado na correção de um grande número de redações em um curto espaço de tempo. A maior parte dos trabalhos se concentra no esforço de desenvolver algoritmos que sejam capazes de corrigir automaticamente textos em inglês. No entanto, para a língua portuguesa, essa ainda é uma área que está em desenvolvimento. Neste contexto, este artigo apresenta um Mapeamento Sistemático da Literatura que busca identificar as abordagens de Inteligência Artificial que estão sendo utilizadas para oferecer suporte à avaliação de redações escritas na língua portuguesa. Os principais achados deste artigo incluem os seguintes fatos: (i) as abordagens dos trabalhos selecionados costumam focar no uso de atributos extraídos do texto em vez do uso de modelos pré-treinados baseados em Deep Learning; (ii) existe prevalência de métricas tradicionais, como Precisão, Cobertura e F-Measure na validação dos resultados; (iii) os feedbacks gerados pelas abordagens possuem um baixo detalhamento; e (iv) os artigos selecionados não analisam o impacto prático em aplicações do mundo real.

A adoção de Learning Analytics (LA) em Instituições de Ensino Superior (IES) é considerada uma tarefa desafiadora, por sugerir mudanças na estrutura da IES e abranger atividades em diversos setores do sistema de ensino. Compreender o nível de maturidade dos projetos de LA existentes dentro das IES torna-se fundamental para que as instituições consigam ampliar seus projetos e abranger todos os seus setores. Este trabalho apresenta um estudo exploratório sobre o nível de maturidade da LA nas IES das regiões do sul e sudeste brasileiro de acordo com o modelo de maturidade MMALA. Durante o estudo foi possível notar que as IES possuem níveis de maturidade diferentes para cada área do processo do MMALA.

With the emergence of generative artificial intelligence, an increasing number of individuals and organizations have begun exploring its potential to enhance productivity and improve product quality across various sectors. The field of education is no exception. However, it is vital to notice that artificial intelligence adoption in education dates back to the 1960s. In light of this historical context, this white paper serves as the inaugural piece in a four-part series that elucidates the role of AI in education. The series delves into topics such as its potential, successful applications, limitations, ethical considerations, and future trends. This initial article provides a comprehensive overview of the field, highlighting the recent developments within the generative artificial intelligence sphere.

A utilização de Learning Analytics (LA) traz consigo diferentes benefícios às instituições de ensino. Porém, exige recursos computacionais e de internet inacessíveis às populações de baixa renda, tornando esta uma tecnologia que pode gerar desigualdade. Nesse contexto, este artigo tem dois objetivos: (i) apresentar o conceito de LA Desconectada, que permite a aplicação dessa tecnologia em ambientes com recursos limitados; e (ii) apresentar uma aplicação real para correção de produção textual de alunos de escolas públicas brasileiras, compatível com este conceito. O aplicativo permite a correção offline de redações escritas no papel e apresenta um dashboard impresso e com informações sumarizadas aos professores.

Learning Analytics (LA) visa a análise de dados educacionais para melhorar o processo de ensino e aprendizagem. Para sua efetiva adoção, é essencial considerar a opinião dos stakeholders. Assim, este artigo tem por objetivo conhecer as expectativas dos estudantes de uma Instituição de Ensino Superior (IES) pública brasileira sobre o uso de seus dados, com o objetivo final de propor diretrizes para a definição de pol´ıticas que apoiem a adoção de LA e atendam às expectativas desses estudantes. Para isso, conduziu-se um estudo de caso com a utilização do questionário do projeto SHEILA para coleta de dados; a análise de dados foi realizada por meio de técnicas estatísticas e Mineração de Dados Educacionais (MDE).

The Jornadas de Atualização em Informática na Educação (JAIE) are moments and contents of a scientific and technological update for the Informatics in Education (IE) community in Brazil. They are organized annually at the Brazilian Congress of Informatics in Education (CBIE) in the form of a set of short courses (tutorials). Each day focuses on current and relevant IE topics. They aim to stimulate the training of researchers in the areas of Computing, Education, Psychology, Design, and other similar areas with contemporary and cutting edge theoretical and methodological approaches. In this JAIE edition, five chapters summarize the state of the art or technique and help to evolve the area: "Advances in Collaborative Learning with Computer Support in Education 4.0", "Creative Computing with Scratch, Mixly and Arduino: Prototyping with HackEduca Conecta”, “Developing Innovative and Exciting Digital Educational Games with the PlayEduc Framework”, “Demystifying the adoption of Learning Analytics: a concise guide on tools and instruments”, and “Analysis of Discussions in Educational Forums Using Text Mining and Graph Analysis”.

Learning Analytics is a new field in education whose adoption can bring benefits for teaching and learning processes. However, many higher education institutions may not be ready to start using learning analytics due to challenges such as organizational culture, infrastructure, and privacy. In this context, Maturity Models (MMs) can support institutions to systematize their processes, enabling them to progress successively in the learning analytics adoption. MMs are used in different fields to support the improvement of processes, describing them in terms of maturity levels, and identifying enhancements that could lead an organization to higher levels of such maturity. Thus, this paper presents an outline of a MM for Learning Analytics adoption in higher education institutions, describing its levels and areas, together with its development methodology.

Learning Analytics (LA) aims to analyze the data generated by both students and teachers in online environments in order to promote actions to improve teaching and learning processes. The results of such analyzes can help teachers to know their students’ study processes, as well as being able to assist with the verification and correction of both educational activities and practices. For students, LA can help with reflection and self-regulation of learning. However, despite its benefits, institutions have difficulties in adopting it. In this sense, an instrument that can support the use of LA is the Maturity Model (MM), which has been used in different knowledge areas in order to indicate an improvement roadmap for organizations. Hence, this paper aims to present the assessment results of a MM proposed for the adoption of LA in Higher Education Institutions, called MMALA. The evaluation focused on the model composition and was carried out through a questionnaire addressed to LA researchers and professionals. After conducting analyzes, both qualitative and quantitative, suggestions for improvement for the proposed model were identified, and the model was validated, supporting its further development.

Learning Analytics (LA) is concerned with the use of data resulting from students' interaction with Learning Management Systems (LMS) to carry out analyses that can help students and teachers to understand and evaluate the learning process. However, several works on the subject only expose future expectations of what LA applications can perform. Thus, this paper aims at finding out practical results of LA and if its use has improved teaching and learning.

Context: Learning Analytics (LA) is a recent trend in education, and its impacts are not widely demonstrated yet. Aim: Identify and summarize practical results of using LA concerning teaching and learning. Methodology: Systematic Literature Review. Results: From 757 papers, 14 have shown evidence to understand practical results for both teachers and students while using LA. Conclusions: LA has contributed to improve teaching and learning in different circumstances, after students receiving automated feedback, as well as has supported teachers to achieve insights about assessments, which could never be revealed without using LA. Resumo. Contexto: Learning Analytics (LA) é uma tendência recente na educação e os impactos de sua adoção ainda não estão amplamente demonstrados. Objetivo: Identificar e sumarizar os resultados práticos da utilização de LA no que se refere ao ensino e aprendizagem. Metodologia: Revisão Sistemática de Literatura. Resultados: De um total de 757 artigos, 14 apresentaram evidências que permitiram entender os resultados práticos do uso de LA para professores e estudantes. Conclusão: LA contribuiu para melhorar o ensino e aprendizagem em diferentes circunstâncias, após os estudantes receberem feedbacks automatizados, e apoiou os instrutores a alcançarem percepções que, sem o apoio de LA, sequer seriam conhecidas.

Due to the increasing amount of produced information, caused by the Web 2.0, the new apps and even the older ones demand a higher degree of scalability, concurrence, and availability, amongst other things. The database models already existing, especially the most used, the relational, could not stand this new paradigm of data manipulation. From this new demand, the NoSQL movement has emerged and it has as its goal to solve problems faced by other models. The right manipulation of this huge quantity of data may lead companies to an improvement in their products and services. However, it is necessary to know which NoSQL database fits better to each one own scenario. For that, in this work we have compared characteristics of three models of key-value NoSQL databases: Cassandra, DynamoDB and Redis. Such a comparison is aimed at assisting the understanding on which scenarios the use of such databases are recommended.

Context: The code ownership has influence on various aspects of software development, such as code quality, cooperation and team knowledge. However, there are few studies from the point of view of developers that seek to understand the advantages and disadvantages of code ownership. Goals: to investigate what are the advantages and disadvantages of practicing shared code ownership from the perspective of the software developers. Methodology: A qualitative study was conducted using a semi-structured interviews in three technology companies with different profiles. We conducted 19 interviews, that were audio recorded and then transcribed. We coded the data using qualitative coding techniques. Results: Considering companies' context, we have found six advantages and six disadvantages of using shared code ownership. Five proposition were presented. Conclusion: It is noted that the practice of shared code ownership, as described in theory, is more suitable for a more experienced programmers teams, who are able to understand codes without assistance. However, the adaptation of the practice, in which the author is consulted before any change is done, it is necessary for less experienced teams, who feel unsecure to modify the code cause other errors for the project.