O trabalho propõe um modelo de estrutura de conteúdos educativos para programas interativos de televisão digital. Seu propósito é fornecer referências para processos colaborativos de produção e organização de conteúdo em redes de aprendizagem coligadas a emissoras de televisão educativa e universitária. A elaboração do modelo teve por base uma revisão bibliográfica de teorias, conceitos e práticas, bem como uma pesquisa-ação com grupos de ensino médio e superior. Sua concepção toma como hipótese a carência de modelos de referência nessa área e contempla a coexistência de aspectos formais e informais de educação, a arquitetura e prática das redes sociais e ambientes de aprendizagem, bem como pontos de inovação em relação a modelos tradicionais de comunicação. São definidos termos, atores, eventos, ambientes, categorias, classes e atributos de conteúdo, com indicação de critérios para sua vinculação síncrona ou assíncrona numa programação, além de requisitos e funções computacionais básicas para sua veiculação seletiva, adaptativa e participativa através de mídias digitais. O modelo é apresentado em duas formas: descritiva, por meio de definições, sentenças e quadros de classificação; e visual, por meio de mapas conceituais inter-relacionados. Uma experiência para validação do modelo proposto foi feita mediante a prototipagem de um ambiente televisivo de ensino-aprendizagem de Engenharia de Produção, com emprego de linguagens que compõem o middleware Ginga (NCL e LUA). A aplicação simula uma fábrica virtual com ambientes em três dimensões que dão acesso a vídeos educativos e conteúdos associados. Os resultados das práticas de modelagem e de prototipagem são avaliados a partir de seus objetivos iniciais e perspectivas de aprimoramento. São propostas possíveis extensões e especificações do modelo com vistas a seu aprimoramento e aplicação. Sustenta-se, por fim, que esse uso interativo da televisão digital na educação depende de modelos sistematizados de conteúdo que contemplem a participação comunitária na produção e veiculação de mídias, com o fortalecimento dos instrumentos de aprendizagem para além da comunicação vertical, centralizada e hierarquizada dos canais tradicionais de radiodifusão. / This research proposes a model to educative content structuring for interactive digital television programs. The intent is to provide references for collaborative production processes and content organization in learning networks associated with university and educative television broadcast services. Modelling activity was preceded by critical literature review, considering theories, concepts and practices reported, in parallel with a collaborative action research. The research conception took as hypothesis the lack of reference models applied to this context and considers the coexistence of formal, informal and non-formal learning and educational aspects; architecture and daily practices in social networks and learning environments; and innovations over traditional communication models. The model defines terms, actors, events, environments, as well as content categories, classes and attributes, indicating criteria for their synchronic or asynchronic association in a dynamic television grid. Basic computational requirements and functions for selective, flexible and participative content distribution through convergent digital media were also indicated. The model is presented in two complementary frames. One is descriptive and uses definitions, sentences and structured tables. The other is visual and uses interconnected conceptual maps. An experimental validation of the proposed model was made through the prototyping of a television learning environment oriented to Industrial Engineering education, developed with the use of programming languages (NCL and LUA) according to the brazilian\'s middleware specifications (Ginga). This application simulates a factory in a virtual 3D enviroment leading to educative videos and their associated contents. Modeling and prototyping outcomes are evaluated considering initial objectives and improvement perspectives. Possible modeling extensions and specifications are proposed and discussed targeting its further developments and effective application. The results indicate that interactive use of digital television in education requires systematic content models covering communitarian participation in both media production and distribution processes, in order to enhance learning instruments beyond vertical, hierarchical and centralized communication sustained by traditional broadcast channels.

Diversos esforços vêm sendo realizados por parte do setor produtivo e da academia no sentido de estabelecer mecanismos inovadores que facilitem a incorporação da medição e comparação de desempenho ao processo de decisão das empresas de construção. Dentre estes esforços, destacamse as iniciativas internacionais para desenvolvimento de sistemas de indicadores para comparação de desempenho e clubes de benchmarking. Em particular, estes clubes podem ser considerados como arranjos colaborativos, que fornecem um meio em que organizações podem adquirir e compartilhar conhecimentos do ambiente externo visando à implementação de melhorias incrementais e inovadoras. Entretanto, ainda, são escassos os trabalhos que explicam as condições nas quais os arranjos colaborativos mostram-se eficazes, os seus efeitos em termos de implementação de melhorias e aprendizagem das empresas, e como as práticas compartilhadas ao longo do processo colaborativo são transformadas em novos conhecimentos em diferentes contextos organizacionais. O objetivo geral desta tese foi estabelecer um conjunto de diretrizes para a realização de processos de benchmarking colaborativo visando à implementação de melhorias nas empresas de construção civil. No presente trabalho, foi estudado o processo de aprendizagem colaborativa ocorrido na realização de um Clube de Benchmarking, formado um por um grupo de vinte empresas de construção. Buscou-se, através deste processo, o desenvolvimento conjunto de um sistema de medição para comparação de desempenho e, posteriormente, a troca de experiências e práticas relacionadas aos indicadores e processos gerenciais selecionados pelas empresas participantes. A estratégia de pesquisa utilizada foi a pesquisa-ação, sendo o estudo subdividido em quatro fases: preparação, estudo empírico, estudos complementares e análise final dos dados. O estudo empírico foi realizado em três etapas, sendo o foco da pesquisa refinado ao final de cada nova etapa. Os resultados foram analisados de acordo com duas unidades de análise: o Clube de Benchmarking, por meio do processo colaborativo, e as empresas, através da análise das trajetórias de colaboração e aprendizagem. Os resultados indicaram que o ambiente colaborativo cria oportunidades para as pessoas refletirem e abstraírem sobre os conhecimentos compartilhados, e que esta reflexão estimula as pessoas a implementar melhorias nas suas empresas. A implementação efetiva destas melhorias, por sua vez, depende de um conjunto de capacidades dos gerentes e das empresas participantes para absorver o conhecimento disponibilizado externamente, de modo a entender, disseminar, transformar e usar este novo conhecimento dentro da empresa. As principais contribuições desta tese estão relacionadas a: (a) proposição de uma estrutura de avaliação da eficácia do processo colaborativo em clubes de benchmarking; (b) compreensão do processo de aprendizagem nas empresas de construção induzido por um processo de benchmarking colaborativo; e (c) o estabelecimento de um conjunto de diretrizes para a realização de processos de benchmarking colaborativo, incluindo os requisitos necessários para o seu desenvolvimento, bem como as principais etapas e passos para a criação de novos grupos. / Several efforts have been developed by practitioner and academics in order to establish innovative mechanisms which facilitate the insertion of performance measurement and comparison in the construction companies decision process. One of these efforts refers to the international initiatives for the development of performance measurement benchmarking systems and benchmarking clubs. In particular, these clubs can be considered as collaborative groups, providing an environment in which companies can acquire and share knowledge from the external environment aiming to implement performance improvements. However, little is known about the conditions under which effective networking takes place and the real benefits achieved by organisations regarding to the implementation of improvements and learning. Besides this, there is a lack of understanding about how the shared practices in collaborative groups can be transformed into a new knowledge in different organizational contexts. The aim of this paper is to establish a set of guidelines for the development of a collaborative benchmarking process aiming the implementation of improvements in construction companies. Specifically, a learning collaborative process was investigated, which was developed in a Benchmarking Club, involving twenty construction companies from the State of Rio Grande do Sul-Brazil. Through this process, a performance measurement system for benchmarking was developed, and, afterwards, the exchange of experience and practice related to the measures and managerial processes, previously defined by these companies, was also developed. This study was carried out using an action research strategy, which was divided into four phased: preparation, empirical study, complementary study and final data analysis. The findings were analysed in two phases. Firstly, the three stages of the collaborative process in the Benchmarking Club were analysed, followed by the stories of collaboration and learning of four construction companies involved in the first two stages of this study. As main conclusions, this work identified that the collaborative environment created the opportunity for the construction managers to reflect and abstract about the shared knowledge. Also, this encouraged these managers to start the implementation of improvements in their companies. The level of these improvements were dependent on a set of individual and organisational abilities to absorb the shared knowledge aiming to understand, diffuse, transform and use this knowledge inside of the companies. The main contributions of this study are: (a) the proposition of a structure to evaluate the effectiveness of the collaborative process in benchmarking clubs; (b) the understanding of the learning process in the construction companies due to the collaborative process; and (c) the establishment of a set of guidelines for the development of a collaborative benchmarking process, including the necessary requirement for its development, as well as the main stages and steps to create new benchmarking groups.

Kyoto University (京都大学) / 0048 / 新制・課程博士 / 博士(工学) / 甲第15350号 / 工博第3229号 / 新制||工||1486(附属図書館) / 27828 / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 小林 潔司, 教授 川﨑 雅史, 教授 藤井 聡 / 学位規則第4条第1項該当

<p>We consider the problem of developing innovative machine learning tools for online education and evaluate their ability to provide instructional resources.  Prediction tasks for student behavior are a complex problem spanning a wide range of topics: we complement current research in student grade prediction and clickstream analysis by considering data from three areas of online learning: Social Learning Networks (SLN), Instructor Feedback, and Learning Management Systems (LMS). In each of these categories, we propose a novel method for modelling data and an associated tool that may be used to assist students and instructors. First, we develop a methodology for analyzing instructor-provided feedback and determining how it correlates with changes in student grades using NLP and NER--based feature extraction. We demonstrate that student grade improvement can be well approximated by a multivariate linear model with average fits across course sections approaching 83\%, and determine several contributors to student success. Additionally, we develop a series of link prediction methodologies that utilize spatial and time-evolving network architectures to pass network state between space and time periods. Through evaluation on six real-world datasets, we find that our method obtains substantial improvements over Bayesian models, linear classifiers, and an unsupervised baseline, with AUCs typically above 0.75 and reaching 0.99. Motivated by Federated Learning, we extend our model of student discussion forums to model an entire classroom as a SLN. We develop a methodology to represent student actions across different course materials in a shared, low-dimensional space that allows characteristics from actions of different types to be passed jointly to a downstream task. Performance comparisons against several baselines in centralized, federated, and personalized learning demonstrate that our model offers more distinctive representations of students in a low-dimensional space, which in turn results in improved accuracy on a common downstream prediction task. Results from these three research thrusts indicate the ability of machine learning methods to accurately model student behavior across multiple data types and suggest their ability to benefit students and instructors alike through future development of assistive tools. </p>

No description available.

This research is a formative evaluation of personal learning networks to determine their applicability for professional development in the architecture and design industry. The researcher seeks to find a catalyst toward discipline-wide realization of integrated design practices. This research initiative was spurred by leaders in the field who indicate that a swift transformation to integrated design practice is required in the discipline in order for the practice of architecture to remain effective in today's global economy. The AIA knowledge community has designated this issue a primary focus for professional development, yet innovative solutions for timely and effective knowledge transfer at a discipline-wide scale do not currently exist. Concurrently, there is active research in computer-based organizational learning within the social sciences, education and the human computer interaction disciplines, indicating its potential as an effective method for the dissemination of knowledge. The research strategy draws upon the human computer interaction discipline's user-centered design philosophy to harness the disciplines knowledge by actively engaging experts in a formative evaluation of personal learning networks using the Delphi method. / Ph. D.

Overall, 11% of high school students leave school without a diploma, and the percentage is higher for at-risk populations. High school graduates earn higher salaries and are healthier and more law abiding than dropouts. Research is limited on the motivation of at-risk students to graduate from high school related to their technological identity to include technology and social media in their learning schema. This qualitative case study explored at-risk students' perceptions of social media, personal learning networks, and informal learning in facilitating their graduation. Pink's concept of motivation, Siemens's connectivism theory, and Bingham and Conner's theory of engagement and social learning provided the conceptual framework. Interviews were conducted with 11 at-risk students identified by one Charter school: 4 students at-risk of dropping out, 3 dropouts planning to return, and 4 dropouts who had returned to high school. Open coding was used to identify rich themes and patterns that may help at-risk students succeed in school. Of the 5 themes identified 4 related to technology identity: transference to learning, relationships with personal learning communities and social networks, bridging technologies, and connected knowledge. Relationships with instructors and the school community also emerged as a theme. Connecting familiar and accessible technologies with formal learning could provide additional means of supporting academic success. Permitting the use of smart phones and social media to provide technological access to learning materials and instructors may create a motivating learning environment where students are willing to remain in high school to obtain a degree. Potential social and work benefits beyond high school may accrue for students.

Aortic Valve Analysis and Area Prediction using Bayesian Modeling Miheer S. Ghotikar ABSTRACT Aortic valve stenosis affects approximately 5 out of every 10,000 people in the United States. [3] This disorder causes decrease in the aortic valve opening area increasing resistance to blood flow. Detection of early stages of valve malfunction is an important area of research to enable new treatments and develop strategies in order to delay degenerative progression. Analysis of relationship between valve properties and hemodynamic factors is critical to develop and validate these strategies. Porcine aortic valves are anatomically analogous to human aortic valves. Fixation agents modify the valves in such a manner to mimic increased leaflet stiffness due to early degeneration. In this study, porcine valves treated with glutaraldehyde, a cross-linking agent and ethanol, a dehydrating agent were used to alter leaflet material properties. The hydraulic performance of ethanol and glutaraldehyde treated valves was compared to fresh valves using a programmable pulse duplicator that could simulate physiological conditions. Hydraulic conditions in the pulse duplicator were modified by varying mean flow rate and mean arterial pressure. Pressure drops across the aortic valve, flow rate and back pressure (mean arterial pressure) values were recorded at successive instants of time. Corresponding values of pressure gradient were measured, while aortic valve opening area was obtained from photographic data. Effects of glutaradehyde cross-linking and ethanol dehydration on the aortic valve area for different hydraulic conditions that emulated hemodynamic physiological conditions were analyzed and it was observed that glutaradehyde and ethanol fixation causes changes in aortic valve opening and closing patterns. Next, relations between material properties, experimental conditions, and hydraulic measures of valve performance were studied using a Bayesian model approach. The primary hypothesis tested in this study was that a Bayesian network could be used to predict dynamic changes in the aortic valve area given the hemodynamic conditions. A Bayesian network encodes probabilistic relationships among variables of interest, also representing causal relationships between temporal antecedents and outcomes. A Learning Bayesian Network was constructed; direct acyclic graphs were drawn in GeNIe 2.0ʾ using an information theory dependency algorithm. Mutual Information was calculated between every set of parameters. Conditional probability tables and cut-sets were obtained from the data with the use of Matlabʾ. A Bayesian model was built for predicting dynamic values of opening and closing area for fresh, ethanol fixed and glutaradehyde fixed aortic valves for a set of hemodynamic conditions. Separate models were made for opening and closing cycles. The models predicted aortic valve area for fresh, ethanol fixed and glutaraldehyde fixed valves. As per the results obtained from the model, it can be concluded that the Bayesian network works successfully with the performance of porcine valves in a pulse duplicator. Further work would include building the Bayesian network with additional parameters and patient data for predicting aortic valve area of patients with progressive stenosis. The important feature would be to predict valve degenration based on valve opening or closing pattern.

The post-9/11 security environment is one that confronts the United States, and specifically the U.S. Army, with complex problems that require development of leaders with improved knowledge, skills, and attributes to meet the challenges of defending the nation. The problem confronting the U.S. Army is that Army leaders lack a learning environment or methodology that enables effective self-development throughout their career. Research suggests that PLNs (PLNs) and a connectivist approach may address that problem, but these have been limited to civilian educational environments. This qualitative case study explored the use of PLNs with a connectivist approach for leader self-development in Army organizations. The case in this study was that of a cohort of 22 officers in a U.S. Army unit. Each officer developed a PLN and then employed a connectivist approach to connect, curate, create, and share knowledge. Data were collected using semistructured interviews and analyzed with a focus on themes that could inform future decisions by Army senior leadership on the self-development of leaders, and that could enable decisions that will reduce loss of life and destruction in wartime. The major themes developed in this study account for the approach and challenges to Army self-development, the motivation for and execution of connected learning, and the structuring and desired characteristics of self-development that employs PLNs and connectivism. Recommendations included integrating a semistructured approach to self-development into emerging educational approaches to Army leader development. More prepared Army leaders can respond to crises more effectively, reducing adverse effects, damage, and loss of life

Mentoring, coaching, and teacher induction programs continue to gain traction in school jurisdictions across Canada and internationally in an effort to address teacher attrition, support professional growth, and improve teaching and learning. Conceptualized as a patchwork quilt, this thesis by article reports on a qualitative case study that examined the Teacher Induction Program (TIP) implemented in 2009 in the Western Québec School Board (WQSB). Each manuscript or fabric block highlights a different component of the TIP that when stitched together reveals an emerging pattern of how professional learning and development, mentoring and coaching, and teacher evaluation is understood in the school district. Anchored within a social learning theoretical framework, this study asked two guiding questions: 1) What is the influence of the mentor–coach role on experienced teachers’ professional learning, practice, and well-being? And 2) How do the WQSB stakeholders perceive the impact of the TIP? Using a single case study design, data were collected from a variety of sources and stakeholder perspectives, including documents, questionnaires, semi-structured interviews, and focus groups. Using the Framework Method, data were abductively analyzed with and against the study’s theoretical and conceptual frameworks. The study’s findings indicate that the mentor–coach role has powerful potential to not only support experienced teachers’ practice-based professional learning, but also to cultivate their sense of well-being and flourishing in schools. However, tensions emerged around how different stakeholders perceived the TIP’s impact in the school district. In particular, further attention is required to clarify the purpose and process of mentoring and coaching as well as the role of teacher evaluation within teacher induction. Ultimately, this study highlights the important role an effective mentoring and coaching fellowship can play in supporting veteran teacher professional learning, practice and well-being, which then has the potential to cultivate positive transformational change in a school district.