The analysis and automatic affect estimation system from human expression has been acknowledged as an active research topic in computer vision community. Most reported affect recognition systems, however, only consider subjects performing well-defined acted expression, in a very controlled condition, so they are not robust enough for real-life recognition tasks with subject variation, acoustic surrounding and illumination change. In this thesis, an artificial intelligence system is proposed to continuously (represented along a continuum e.g., from -1 to +1) estimate affect behaviour in terms of latent dimensions (e.g., arousal and valence) from naturalistic human expressions. To tackle the issues, feature representation and machine learning strategies are addressed. In feature representation, human expression is represented by modalities such as audio, video, physiological signal and text modality. Hand- crafted features is extracted from each modality per frame, in order to match with consecutive affect label. However, the features extracted maybe missing information due to several factors such as background noise or lighting condition. Haar Wavelet Transform is employed to determine if noise cancellation mechanism in feature space should be considered in the design of affect estimation system. Other than hand-crafted features, deep learning features are also analysed in terms of the layer-wise; convolutional and fully connected layer. Convolutional Neural Network such as AlexNet, VGGFace and ResNet has been selected as deep learning architecture to do feature extraction on top of facial expression images. Then, multimodal fusion scheme is applied by fusing deep learning feature and hand-crafted feature together to improve the performance. In machine learning strategies, two-stage regression approach is introduced. In the first stage, baseline regression methods such as Support Vector Regression are applied to estimate each affect per time. Then in the second stage, subsequent model such as Time Delay Neural Network, Long Short-Term Memory and Kalman Filter is proposed to model the temporal relationships between consecutive estimation of each affect. In doing so, the temporal information employed by a subsequent model is not biased by high variability present in consecutive frame and at the same time, it allows the network to exploit the slow changing dynamic between emotional dynamic more efficiently. Following of two-stage regression approach for unimodal affect analysis, fusion information from different modalities is elaborated. Continuous emotion recognition in-the-wild is leveraged by investigating mathematical modelling for each emotion dimension. Linear Regression, Exponent Weighted Decision Fusion and Multi-Gene Genetic Programming are implemented to quantify the relationship between each modality. In summary, the research work presented in this thesis reveals a fundamental approach to automatically estimate affect value continuously from naturalistic human expression. The proposed system, which consists of feature smoothing, deep learning feature, two-stage regression framework and fusion using mathematical equation between modalities is demonstrated. It offers strong basis towards the development artificial intelligent system on estimation continuous affect estimation, and more broadly towards building a real-time emotion recognition system for human-computer interaction.

The intent of this thesis is to guide further research and discussion of C-Print, meaning-for-meaning transcription and its applications to today’s dynamic classroom settings under a Universal Design Paradigm. Evidence suggests that providing these captions can benefit Deaf and Hard of Hearing populations and also that concise, textual representations of information increase retention for average learners in multimedia settings. Individual differences were considered and low internal control participants did significantly better on exams when material was captioned compared to when it was not. They also tended to outperform high internal control participants on captioned material exams.

Localizado na modalidade de aprendizagem um computador por aluno e no contexto da proposta metodológica de PAs (Projetos de Aprendizagem), este estudo tem por objetivo apresentar uma dinâmica de acompanhamento da conceituação a partir da enunciação sobre as produções escritas dos alunos. Parte-se da concepção de enunciação como apropriação ativa da língua, na qual se expõe a compreensão em muitos casos ausente no registro textual. Ferramentas do laptop em utilização na escola pólo do pré-piloto do Projeto UCA (Um Computador por Aluno) em Porto Alegre e o ambiente virtual de aprendizagem Amadis dão suporte ao trabalho. Sustentam a análise o Método Clínico de Piaget e a teoria de reações compensadoras proposta pelo mesmo autor, aplicáveis à amplitude de temas contemplados nos PAs e aos diferentes percursos empreendidos pelos estudantes em suas pesquisas. / Located in 1:1 (one computer per student) learning modality and in the context of Learning Projects methodology, this study aims to present a dynamics of monitoring the process of conceptualization considering students’ written productions. For the purpose of this study, enunciation is understood as an active process of language ownership, which exposes comprehension often lacking in textual speech. Resources from the laptop used in the trial school of UCA (One Computer per Child) Project pre-pilot in Porto Alegre and the virtual learning environment Amadis support the work. The analysis is based on Jean Piaget Clinical Method and on the Cognitive Compensation Reactions theory, proposed by the same author, applicable to the variety of themes covered in the projects and to the different paths taken by students in their researches.

Localizado na modalidade de aprendizagem um computador por aluno e no contexto da proposta metodológica de PAs (Projetos de Aprendizagem), este estudo tem por objetivo apresentar uma dinâmica de acompanhamento da conceituação a partir da enunciação sobre as produções escritas dos alunos. Parte-se da concepção de enunciação como apropriação ativa da língua, na qual se expõe a compreensão em muitos casos ausente no registro textual. Ferramentas do laptop em utilização na escola pólo do pré-piloto do Projeto UCA (Um Computador por Aluno) em Porto Alegre e o ambiente virtual de aprendizagem Amadis dão suporte ao trabalho. Sustentam a análise o Método Clínico de Piaget e a teoria de reações compensadoras proposta pelo mesmo autor, aplicáveis à amplitude de temas contemplados nos PAs e aos diferentes percursos empreendidos pelos estudantes em suas pesquisas. / Located in 1:1 (one computer per student) learning modality and in the context of Learning Projects methodology, this study aims to present a dynamics of monitoring the process of conceptualization considering students’ written productions. For the purpose of this study, enunciation is understood as an active process of language ownership, which exposes comprehension often lacking in textual speech. Resources from the laptop used in the trial school of UCA (One Computer per Child) Project pre-pilot in Porto Alegre and the virtual learning environment Amadis support the work. The analysis is based on Jean Piaget Clinical Method and on the Cognitive Compensation Reactions theory, proposed by the same author, applicable to the variety of themes covered in the projects and to the different paths taken by students in their researches.

Localizado na modalidade de aprendizagem um computador por aluno e no contexto da proposta metodológica de PAs (Projetos de Aprendizagem), este estudo tem por objetivo apresentar uma dinâmica de acompanhamento da conceituação a partir da enunciação sobre as produções escritas dos alunos. Parte-se da concepção de enunciação como apropriação ativa da língua, na qual se expõe a compreensão em muitos casos ausente no registro textual. Ferramentas do laptop em utilização na escola pólo do pré-piloto do Projeto UCA (Um Computador por Aluno) em Porto Alegre e o ambiente virtual de aprendizagem Amadis dão suporte ao trabalho. Sustentam a análise o Método Clínico de Piaget e a teoria de reações compensadoras proposta pelo mesmo autor, aplicáveis à amplitude de temas contemplados nos PAs e aos diferentes percursos empreendidos pelos estudantes em suas pesquisas. / Located in 1:1 (one computer per student) learning modality and in the context of Learning Projects methodology, this study aims to present a dynamics of monitoring the process of conceptualization considering students’ written productions. For the purpose of this study, enunciation is understood as an active process of language ownership, which exposes comprehension often lacking in textual speech. Resources from the laptop used in the trial school of UCA (One Computer per Child) Project pre-pilot in Porto Alegre and the virtual learning environment Amadis support the work. The analysis is based on Jean Piaget Clinical Method and on the Cognitive Compensation Reactions theory, proposed by the same author, applicable to the variety of themes covered in the projects and to the different paths taken by students in their researches.

The introduction of project-based learning into university engineering programs has been shown to positively benefit students that prefer a hands-on experience and give future employers assurance that recent graduates have the tools to handle real-world problems as opposed to theoretical situations. Enhancing the engineering self-efficacy of students, recent graduates and seasoned engineers is made possible through the solution of complex, open-ended problems typically found in engineering design. A high engineering self-efficacy, in turn, positively reflects a person's perception of their complex problem-solving capacity which is critical throughout the design process. The decision to either work virtually with a team or onsite with group members nearby may also further influence self-efficacy and, ultimately, the designer's success. This raises the question explored in this study: Will a design challenge impact engineering self-efficacy equally for online and in-person participants? Two groups engaged in a design challenge to develop a mechanism meant for drone applications, where one group designed and tested their solution in-person, while the other group sent design plans to a third-party for fabrication and testing. Participants filled out a prototype engineering self-efficacy scale before and after the challenge, revealing a significant difference between these two modalities. The small sample size is noted as the cause for inaccuracies and surprising findings. Guidelines for methodology implementation in a larger scale study are included. / Master of Science / In project-based learning courses, students work in groups to make a prototype or other solution to a stated problem, which are helpful for building student confidence in problem-solving, critical thinking and, especially, engineering skills. This confidence translates to believing that carrying out a specific task will lead to success with little-to-no feelings of fear or failure. This generally describes "self-efficacy," and it can apply to any profession. Traditionally, hands-on projects are done in person, where an exchange of ideas is clear and any problems can be handled immediately. However, with schools closed due to the COVID-19 pandemic, these courses had to shift online, leading some to believe that students would not receive the same level and quality of engineering education. Online learning has been around for over 30 years and studies show that students learn just as much, if not more and better, online than sitting in a classroom. Can the same be said for taking part in an engineering project over the internet? Two groups designed a prototype drone attachment, where members of one group worked side-by-side to build and test their solution, while members of the other group worked online and sent files and assembly instructions to a third party. Each participant also filled out a questionnaire before and after the challenge to track their engineering self-efficacy. The limited data led to the conclusion that there is a noticeable difference between the two project completion methods, most likely caused by a low number of participants. The lessons learned from this study were used to create guidelines for a larger-scale study.

No description available.

No description available.

No description available.