The objective of this study is twofold: First, it aims at analyzing and interpreting Brazilian products exportation from 1997 to 2019. Second, it seeks to reduce the scope to the Alagoas state – one of the nation’s poorest states – to identify possible exportation growth opportunities. Thereby, the study was conducted by using the model Product Space, which can show us, through a network, if it is possible to industries produce new products from those already produced by them. After analyzing the Product Space, we concluded that Alagoas is not isolated in the network. Alagoas can grow its network by exporting more than 50 products, such as motor vehicles, chemical wood pulp, and passenger cars. Such growth can promote new public and private partnerships. Thus, new industries might be attracted and boost Alagoas economic growth.

This work presents an interdisciplinary assessment that looks in-depth the tracking of air quality in urban environments. This kind of application is well suited to be approached with the paradigm of wireless sensor networks in their overall variations. Therefore a robust and diverse set of solutions have been developed following the technology capabilities advance.
The proposed experiment takes advantage of Vehicle Sensor Networks (VSN) by embedding sensor nodes to public transportation, addressing this study case with bus lines within such a way that the mobiles spread the sampling activity through a large number of different places visited during the route. Simultaneously, it alleviates restrictions of power management, packaging dimensions (size and weight), and overall maintenance issues. We perform environmental modeling based on real data considering a temporal and spatial multivariate behavior on observed phenomena. We define the city of São Paulo as reference for the case study and parsed the asserted data to create a multivariate map of samples, showing the behavior of five different air pollutants (CO, O3, PM10, NO2 and SO2) simultaneously while it also varies in time.
The current development stage covers a set of handling processes over an input data that has unformatted or missing information due to being sourced from real sensors and the creation of the map mentioned above.
Our methodology addresses
1) the mentioned environmental simulation,
2) deploying mobile sensor nodes and perform sensing process,
3) implement network activity and delivery of collected data,
4) predict a view of a monitored environment based on gathered data with the same strategy previously applied to create the initial field.
Finally, we plan to evaluate system-level performance and operational constraints through an event-based simulation, taking into account a detailed description of roads, bus lines, vehicle itineraries, and overall traffic information.

..

The search for increasingly efficient learning in online environments has encouraged the growing development of research. Learning is often perceived as a tedious activity, which intensifies the very high dropout rates. In recent years, several methods have been applied in order to keep the student engaged in the learning environment in such a way as to transform the tedious activity into a pleasurable and fun one. Computer games allow students to experience ways of learning that demonstrate immersion in a practice, supported by structures that lead to professional experiences and skills, thus, games are seen as a tool that facilitates and supports student learning. In this sense, Gamification uses game elements in non-game environments, with the aim of promoting student engagement, transforming the tedious activity into a pleasurable one. Although several works show positive results with the use of gamification, it is common not to get the student to awaken independence in the learning process. Self-regulation of learning is characterized as the student's ability to act in their learning environment, reflecting and making decisions about which path would be more convenient for better learning. In this way, the student acts in an active and constructive way in the process, in which they establish goals that will guide their learning. In this sense, the OLM (Open Learning Model) enables the opening of the student model, and thus, the student will be able to view information related to the learning process itself, allowing active participation in the learning process, being able to make decisions about which are the paths that generate more results for him in the online environment. Thus, the present work aimed to verify, through a controlled experiment, whether the use of gamification in an online environment with an open model would bring significant results in self-regulation and student learning. For this, two open model environments were compared, one with and the other without gamification, in which students (schools, institutes and public universities) would have to answer 15 questions in an application developed for android, the first 10 being common to all and the other 5 questions customized by them, regarding the level and content of the questions, and at this stage the student would have the version with or without gamification, randomly distributed after accessing the application. The results showed that there was no significant difference between versions with and without gamification, both with regard to self-regulation and learning.

With the development of high violence games, such as: Mortal Combat, Carmageddon, Doom, GTA (Grand Theft Auto), these games are banned in Brazil and in several countries around the world due to their high violence, games in this category pass on the teaching of that to be successful in the game it is necessary to practice unethical attitudes such as: lack of respect, injustices, theft, lack of solidarity, illegal attitudes; emerges as an alternative the Solidary Educational Digital Games (JDES) which aim to place the player in environmental scenarios, catastrophes, having a chance to perform ethical or unethical attitudes being rewarded or punished according to their attitudes. For the development of solidary educational digital games, it is essential not to always start the creation from scratch, beginning, it is essential to use software reuse techniques. This work presents a Conceptual Framework whose purpose is to streamline and cheapen the development of JDES. The Conceptual Framework was divided into two stages, the first was developed through UML modeling (Unified Modeling Language) the Framework Digital Games Educational Solidarity (FJDES) and Auxiliary Classes that were used to create the Solidary Educational Digital Game Flood, in the second step was developed through Agile Modeling and Scrum Method the Scrum Framework Educational Solidarity Digital Games (FSJDES), where the Framework Educational Solidarity Digital Games and Auxiliary Classes already developed in the first step was reused for the creation of the Digital Educational Solidarity Game Earthquake, both games were implemented in Java. As a result, there was a three-month reduction in the development of Solidary Educational Digital Games, where the production of the first game took four months and the second game took a month.

Online learning environments were developed as an online way for students to interact with their teachers, colleagues and course content. These environments have pedagogical resources in quantity and diversity, allowing recommendations from these to students, based on their respective needs, in addition to the skills that should be developed. However, there are limitations in these recommendation systems in online learning environments, especially in relation to the quality of the recommendations, leading to problems regarding the acceptance of the recommendations by the students. One of the factors that can lead to non-acceptance of the recommendations received is the lack of confidence. A possible approach to this problem is to explain to students how this recommendation was created. For this, we intend to identify if there is a difference in relation to the students' confidence in the recommendations: (1) when the recommendation has an explanation coherent with the problem faced by the student; (2) When the recommendation has an explanation not consistent with the problem faced by the student and; (3) When the recommendation is not explained. To evaluate our proposal, we will use simulated data from student interactions in an online course to generate recommendations based on the interactions and needs of the students (simulated). Then, we will create the explanations for the recommendations created and evaluate, using a questionnaire, the effect of the explanation's coherence on the students' confidence in relation to the pedagogical recommendations. We hope with this proposal to contribute to an improvement in the quality of the recommendations and to increase students' confidence in relation to the pedagogical recommendations in online learning environments.

Early identification contributes to language development and other developmental milestones in children with hearing loss. After diagnosis, parents end up having to make several decisions regarding the treatment of their children. In the Brazilian literature, no instrument was found available to provide support for parents' decision process. To address this gap, this work aims to translate and culturally adapt a monitoring screening protocol for parents and professionals that targets children with hearing loss. For this, two studies were conducted. The first consisted of a systematic review of the literature and meta-analysis on the influence of the intervention on the oral language development of children with hearing loss. Fifteen studies were part of the qualitative synthesis and twelve of the quantitative synthesis. According to the summarized data, the age of identification and early intervention allowed children to achieve near-normal speech and language outcomes, both of which can be obtained by growth curves using standardized language scores. The second study consisted of the cultural adaptation and adaptation of 452 items of the Success from the Star protocol, corresponding to all principal areas of language development: communication, listening, listening, and vocalization. Judges evaluated 25% of the items on the following categories: 1) cultural suitability, 2) conceptual clarity, and 3) suitability to the target audience. We proposed a back-translation to consolidate the final items. Subsequently, the translated items were compared with the revised version of the instrument, which comprises 347 items in the two areas of development. A committee of experts qualitatively evaluated the instrument to validate its content. In addition to committee valuation, we conclude with some directions to effectiveness validation through its use by parents and professionals.

Devido aos avanços tecnológicos, várias atividades humanas se transformaram, inclusive na área educacional. Novas abordagens surgiram com o objetivo de promover o uso da tecnologia na educação. Uma dessas abordagens é o e-learning. E-learning envolve a utilização de tecnologias da informação e comunicação para melhorar e ajudar o ensino e a aprendizagem. Devido as recentes pesquisas que apoiam a efetividade de sistemas e-learning, acadêmicos e profissionais estão cada vez mais dando atenção a esse abordagem. Contudo, estado de tédio que levam estudantes a desmotivação é muito persistente nesses sistemas e para lidar com esse problema, gamificação esta sendo aplicada para manter estudantes engajados e motivados durante o processo de aprendizagem. Contudo, e-learning gamificados no estado da arte não monitoram o impacto da gamificação durante o processo de aprendizagem e também não suportam a adaptação do design de gamificação durante a instrução quando a performance dos alunos não está como o esperado, nem de forma automática e nem por humanos. Considerando que alguns estudos reportaram um impacto negativo com o uso da gamificação em sistemas e-learning e que uma das possíveis causas é a falta de adaptação do design da gamificação, uma forma de evitar esses resultados negativos é monitorar e adaptar o design da gamificação quando o design atual não está sendo efetivo para conquistar os objetivos desejados. Portanto, o objetivo desta dissertação é desenvolver uma solução que possibilite professores monitorarem e adaptarem o design de gamificação em e-learning gamificados com o objetivo de manter os estudantes engajados durante todo o processo de aprendizagem e aumentar a percepção positiva dos professores em relação aos efeitos da gamificação. A solução apresentada será implementada e complementará uma ferramenta de autoria que permite a criação e customização por professores de ambientes educacionais gamificados já existente. Consequentemente, a nova versão da ferramenta de autoria permitirá professores monitorarem a efetividade da gamificação e adaptar o cenário gamificado durante o processo de aprendizagem, quando julgarem necessário. Após a expansão das funcionalidades da ferramenta de autoria, objetiva-se avaliar a solução e apresentar as vantagens, desvantagens, limitações e possíveis trabalhos futuros. 

A busca pela alocação efetiva dos recursos a fim de alcançar o maior número de pessoas é o desafio constante do gestor público. Desta forma, o presente trabalho tem como objetivo identificar quais fatores, percebido pelos gestores educacionais, impacta no IDEB das escolas municipais de Maceió-Alagoas. Os dados iniciais trabalhados nesta pesquisa já existem, porém precisam ser analisados frente à realidade cotidiana do âmbito escolar, estabelecendo uma relação entre a percepção dos gestores e os resultados divulgados pelo INEP/MEC. Assim, surge a pergunta central que norteia esse estudo:  comointegrar a visão computacional (máquina) com a visão dos gestores educacionais (humanos) para identificar os fatores que possam influenciar positivamente a melhoria do IDEB, visando apoiar a tomada de decisão dos gestores educacionais? Para tanto, utilizou-se dados do teste SAEB das escolas públicas de Maceió. A presente pesquisa tem caráter descritivo e o desenvolvimento metodológico da investigação advém da revisão da literatura, da análise e compilação dos dados do rendimento dos alunos dos anos iniciais do ensino fundamental da rede de ensino municipal de Maceió-Alagoas no IDEB do ano de 2015, e da realização de entrevistas com os diretores escolares, visando compreender as impressões da gestão escolar sobre os resultados do IDEB, conduzindo um estudo experimental, obtendo resultados na tarefa de identificação dos atributos relevantes, como os extraídos pela visão computacional principalmente nas dimensões: violência na escola; políticas, ações e programas na escola; gestão escolar e recursos financeiros.

Modelos de linguagem pré-treinados vêm ganhando notoriedade e resultados promissores. Inicialmente existiam duas grandes limitações nessa abordagem, a primeira é que o treinamento de modelos existentes demandava grande quantidade de recursos computacionais, o que podia inviabilizar sua utilização, e a segunda seriam os modelos multilíngues, que fatidicamente tinham baixo desempenho em linguagens de baixo recurso justamente pela sua alta generalidade. Esses dois pontos focais foram mitigados com a utilização de multi modelos com técnicas de ajuste fino em datasets para idiomas específicos. Contudo, ainda existem processos manuais na seleção de alguns hiperparâmetros e por isso propomos uma abordagem genética, a fim de alcançar a solução de forma mais automatizada e prática para pesquisadores e professionais da área. Para tal, foram avaliados dois conjuntos de dados para classificação, alcançando resultados equivalentes ou superiores aos encontrados em trabalhos anteriores que utilizam uma abordagem de seleção manual de parte dos hiper-parâmetros.

Systemic Lupus Erythematosus (SLE) is a chronic, autoimmune, multisystemic inflammatory disease, whose changes in the central nervous system (CNS) can be assessed by Magnetic Resonance Imaging (MRI). The aim of this study is to propose a model of brain MRI interpretation based on correlations between neuroimaging changes and clinical and laboratory data. Brain MRI of 44 patients were analised in this cross-sectional study, considering the variables: brain atrophy, white matter hiperintense lesion (WMHI), arterial stenoses, lacunar infarcts and micro-bleeds. Proton Spectroscopy and Diffusion Tension Imaging (DTI) were also performed, with Fractional Anisotropy (FA) and Medium Diffusion (MD) analysis. Among the clinical variables disease time, disease activity index (SLEDAI) and cumulative organic damage (SLICC) were considered, as well as the presence of SLE criteria by the American College of Rheumatology (ACR). Data analysis verified normal distribution, followed by Pearson's Correlation, T-Test and Chi-square. The sample included 43 women and one man, with an average age of 34.4 ± 11.3; mean illness time of 51 months ± 42.3; SLEDAI average 3.37 ± 3.6, SLICC average 0.37 ± 0.54. Mild brain atrophy was identified in 40.9% of patients, while WMHI was present in 65.9%. There were stenoses and micro-bleeding in 13.6% of the sample and in 15.9% lacunar infarcts were identified. Average NAA / Creatine ratio of 1.94 ± 0.15, Mio / Creatine ratio of 0.20 ± 0.04. The changes in FA and MD were associated with age, disease duration and global brain atrophy score. There was an association between ACR criteria and changes in FA and MD. Correlation of increased WMHI with age and with the presence of arterial stenosis was evidenced. The average time of illness, SLEDAI and SLICC were reduced, inferring that the sample was mostly in the period of remission of the disease. MRI findings were similar to the literature when WMHI, atrophy, stenosis, microsleeds and lacunas infarcts were evaluated. Microstructural changes associated with ACR criteria for SLE were evidenced. In contrast, we found no statistically significant correlation between FA and MD with SLEDAI and SLICC. In view of the subtlety and importance of MRI findings, a model of brain MRI interpretation in SLE patients is recommended.

A catarata é uma doença ocular que se trata de qualquer opacificação do cristalino que atrapalhe a entrada de luz nos olhos, ou seja, há a perda de transparência do cristalino, acarretando na diminuição da visão. Segundo a Organização Mundial de Saúde (OMS), a catarata é responsável por cerca de 47,8% dos casos de cegueira no mundo, acometendo principalmente a população idosa. A deficiência visual causada pela catarata pode ser revertida com o tratamento cirúrgico, o qual consiste na remoção da lente natural opaca e colocação de uma lente artificial, chamada de lente intraocular. Apesar da cirurgia de catarata ser o procedimento cirúrgico mais realizado na oftalmologia e uma das que mais evoluiu nas últimas décadas, trata-se de um procedimento microscópico de alta complexidade, muito seguro, porém, como qualquer procedimento invasivo, tem os seus riscos associados. A tecnologia atual e a experiência do cirurgião reduzem significativamente esse risco. A evolução conseguida foi tamanha que a cirurgia deixou de ser somente reparativa para ser também refrativa, ou seja, tendo também o objetivo de corrigir o erro refrativo (grau) da paciente. Nesse contexto, uma das questões mais críticas é a escolha da lente ideal para o paciente, pois caso ocorra algum equívoco nessa etapa, pode culminar em uma insatisfação por parte do paciente, no caso se ainda precisar corrigir muito o grau pelo uso de óculos ou até mesmo, em casos mais graves, a necessidade de uma outra cirurgia para que a satisfação seja conseguida. Por se tratar de um procedimento delicado e custoso, a escolha da lente se trata de um tomada de decisão sensível e importante, pois deve-se levar em conta que é uma situação muito complicada para o paciente se submeter a tal procedimento e vir a ter complicações pós cirurgia devido à dificuldade de qualquer reparo posterior no local afetado. Para auxiliar o profissional no momento da escolha da lente adequada, esse estudo propõe a construção de um modelo preditivo que irá sugerir o modelo de lente de mais adequado para cada paciente. O objeto do estudo terá como base um banco de dados contendo aproximadamente cinco mil pacientes que foram submetidos ao procedimento cirúrgico da catarata. Essa base foi construída a partir da coleta de dados dos prontuários pré e pós cirúrgicos de todos os pacientes disponíveis. A base será submetida à uma análise estatística para justificar e confirmar a significância de cada atributo presente na base, viabilizando assim uma seleção de variáveis que serão utilizados como entrada para os algoritmos de aprendizagem. O resultado esperado dessa metodologia consiste em uma modelagem computacional que proporcione a concepção de um modelo preditivo que tenha a capacidade de propor um modelo de lente o mais adequado possível, fazendo com que se eleve a assertividade e a satisfação dos resultados dos procedimentos cirúrgicos praticados atualmente.