Thesis (M.S.)--Indiana University, 2003. / Includes bibliographical references (leaves 74-79).

Thesis (Ph. D.)--Oregon State University, 2007. / Printout. Includes bibliographical references (leaves 332-337). Also available on the World Wide Web.

Thesis (Ph. D.)--University of California, San Diego, 2007. / Title from first page of PDF file (viewed June 11, 2007). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 187-195).

We start by showing that in an active learning setting, the Perceptron algorithm needs $\Omega(\frac{1}{\epsilon^2})$ labels to learn linear separators within generalization error $\epsilon$. We then present a simple selective sampling algorithm for this problem, which combines a modification of the perceptron update with an adaptive filtering rule for deciding which points to query. For data distributed uniformly over the unit sphere, we show that our algorithm reaches generalization error $\epsilon$ after asking for just $\tilde{O}(d \log \frac{1}{\epsilon})$ labels. This exponential improvement over the usual sample complexity of supervised learning has previously been demonstrated only for the computationally more complex query-by-committee algorithm.

Thesis (Ed.D.)--Duquesne University, 2005. / Title from document title page. Abstract included in electronic submission form. Includes bibliographical references (p. 99-102) and index.

Orientador: Odete Pacubi Baierl Teixeira / Banca: Sandra Regina Teodoro Gatti / Banca: Rui Manoel de Bastos Vieira / Resumo: Esta pesquisa, de cunho, qualitativo, buscou investigar como alunos no nono ano fornecem explicações relacionadas aos processos de eletrização quando deparados com atividades interativas, objetivando responder quais componentes estão presentes nos argumentos. Participaram do trabalho uma professora em exercício no Ensino Fundamental - que também esteve no papel de pesquiadora - e quatro alunos do nono ano de uma escola particular no interior de São Paulo. Foram propostos seis problemas: dois sobre eletrização por atrito; dois sobre eletrização por contato e dois sobre indução e polarização, sendo utilizados com os alunos nesta sequência. A aplicação das atividades ocorreu em três etapas. Na primeira, foi apresentado em problema maior que tratava sobre choques elétricos, seguido de um tempo para que os alunos registrassem, individualmente em uma folha, a resolução e explicação para este fato. A segunda etapa consistiu na aplicação das seis atividades, sempre se iniciando com o registro individual sobre a hipótese criada para a solução de cada problema, seguido da validação ou não da hipótese, assim como, a reescrita do modelo explicativo. Finalmente, a terceira etapa, consistiu na retomada do problema maior, objetivando verificar, ou não, alguma mudança sobre a primeira explicação fornecida pelos alunos. A análise dos dados foi elaborada partindo de três momentos. O primeiro momento foi referente ao problema maior. O segundo momento esteve relacionado, particularmente, as seis situações vivenciadas pelos alunos envolvendo a eletrização. E o terceiro momento tratou-se da retomada do problema maior, após a realização das seis atividades. Os dados analisados demonstram que este tipo de atividade permite que os alunos comecem a estruturar algumas relações referentes à eletrização, sendo válidas também para aulas posteriores, em que o professor trabalhará conceitualmente o tema abordado e poderá... / Abstract: This research, of qualitative nature, aimed to investigate how ninth grade students provide explanations related to the electrification processes using interactive activities, aiming to answer the components present in the arguments. The participants in the work were a middle shool teacher, also the researcher, and four ninth grade students from a private school in the state of São Paulo. Six problems were proposed: two about electrification by friction; two about electrification by contact and two about induction and polarization, all done by the students in this order. The application of these activities occurred in three steps. First, a larger problem of shock was presented, followed by a time for the students to individually record results about the solution for and explanation of the issues. The second step was the application of the six activities, always starting from the individual results about the hypothesis creating a solution for each problem, followed by the validation or invalidation of the hypothesis, as well as rewriting the exploratory model. The third and final stage restated the larger problem in order to verify whether or not any changes needed to be made in the first explanation provided by the students. Data analysis was prepared three times. The first time was related to the larger problem. The second time was particularly related to the six problems introduced to the students about electrification. The third time dealt with restating the larger problem after the completing the six activities. The data analyzed showed that these types of activities allow students to being making connections related to electrification also valid for future classes, where the teacher will conceptually deal with the topic and use the results of the activities to prepare classes that break down the confusion that was apparent during the practice. The data also showed the relation between student involvement in the activities and their... / Mestre

Made available in DSpace on 2016-02-05T18:29:56Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-05-21. Added 1 bitstream(s) on 2016-02-05T18:34:01Z : No. of bitstreams: 1 000855155_20170818.pdf: 488200 bytes, checksum: fc5b076cad41e4aa343e994401d138ac (MD5) Bitstreams deleted on 2017-08-18T12:37:06Z: 000855155_20170818.pdf,. Added 1 bitstream(s) on 2017-08-18T12:37:51Z : No. of bitstreams: 1 000855155.pdf: 3951646 bytes, checksum: de764ce7b205b6ab8017443d7162d10a (MD5) / Esta pesquisa, de cunho, qualitativo, buscou investigar como alunos no nono ano fornecem explicações relacionadas aos processos de eletrização quando deparados com atividades interativas, objetivando responder quais componentes estão presentes nos argumentos. Participaram do trabalho uma professora em exercício no Ensino Fundamental - que também esteve no papel de pesquiadora - e quatro alunos do nono ano de uma escola particular no interior de São Paulo. Foram propostos seis problemas: dois sobre eletrização por atrito; dois sobre eletrização por contato e dois sobre indução e polarização, sendo utilizados com os alunos nesta sequência. A aplicação das atividades ocorreu em três etapas. Na primeira, foi apresentado em problema maior que tratava sobre choques elétricos, seguido de um tempo para que os alunos registrassem, individualmente em uma folha, a resolução e explicação para este fato. A segunda etapa consistiu na aplicação das seis atividades, sempre se iniciando com o registro individual sobre a hipótese criada para a solução de cada problema, seguido da validação ou não da hipótese, assim como, a reescrita do modelo explicativo. Finalmente, a terceira etapa, consistiu na retomada do problema maior, objetivando verificar, ou não, alguma mudança sobre a primeira explicação fornecida pelos alunos. A análise dos dados foi elaborada partindo de três momentos. O primeiro momento foi referente ao problema maior. O segundo momento esteve relacionado, particularmente, as seis situações vivenciadas pelos alunos envolvendo a eletrização. E o terceiro momento tratou-se da retomada do problema maior, após a realização das seis atividades. Os dados analisados demonstram que este tipo de atividade permite que os alunos comecem a estruturar algumas relações referentes à eletrização, sendo válidas também para aulas posteriores, em que o professor trabalhará conceitualmente o tema abordado e poderá... / This research, of qualitative nature, aimed to investigate how ninth grade students provide explanations related to the electrification processes using interactive activities, aiming to answer the components present in the arguments. The participants in the work were a middle shool teacher, also the researcher, and four ninth grade students from a private school in the state of São Paulo. Six problems were proposed: two about electrification by friction; two about electrification by contact and two about induction and polarization, all done by the students in this order. The application of these activities occurred in three steps. First, a larger problem of shock was presented, followed by a time for the students to individually record results about the solution for and explanation of the issues. The second step was the application of the six activities, always starting from the individual results about the hypothesis creating a solution for each problem, followed by the validation or invalidation of the hypothesis, as well as rewriting the exploratory model. The third and final stage restated the larger problem in order to verify whether or not any changes needed to be made in the first explanation provided by the students. Data analysis was prepared three times. The first time was related to the larger problem. The second time was particularly related to the six problems introduced to the students about electrification. The third time dealt with restating the larger problem after the completing the six activities. The data analyzed showed that these types of activities allow students to being making connections related to electrification also valid for future classes, where the teacher will conceptually deal with the topic and use the results of the activities to prepare classes that break down the confusion that was apparent during the practice. The data also showed the relation between student involvement in the activities and their...

abstract: Machine learning methodologies are widely used in almost all aspects of software engineering. An effective machine learning model requires large amounts of data to achieve high accuracy. The data used for classification is mostly labeled, which is difficult to obtain. The dataset requires both high costs and effort to accurately label the data into different classes. With abundance of data, it becomes necessary that all the data should be labeled for its proper utilization and this work focuses on reducing the labeling effort for large dataset. The thesis presents a comparison of different classifiers performance to test if small set of labeled data can be utilized to build accurate models for high prediction rate. The use of small dataset for classification is then extended to active machine learning methodology where, first a one class classifier will predict the outliers in the data and then the outlier samples are added to a training set for support vector machine classifier for labeling the unlabeled data. The labeling of dataset can be scaled up to avoid manual labeling and building more robust machine learning methodologies. / Dissertation/Thesis / Masters Thesis Engineering 2017

Submitted by MARCIO ROBERTO GONÇALVES DE VAZZI null (profmarcio@vazzi.com.br) on 2017-08-23T17:33:30Z No. of bitstreams: 1 Dissertação Marcio 2017_08_23.pdf: 2938261 bytes, checksum: 8035c3e762251cbcf988f6962199a956 (MD5) / Approved for entry into archive by Monique Sasaki (sayumi_sasaki@hotmail.com) on 2017-08-23T19:37:09Z (GMT) No. of bitstreams: 1 vazzi_mrg_me_arafcl.pdf: 2938261 bytes, checksum: 8035c3e762251cbcf988f6962199a956 (MD5) / Made available in DSpace on 2017-08-23T19:37:09Z (GMT). No. of bitstreams: 1 vazzi_mrg_me_arafcl.pdf: 2938261 bytes, checksum: 8035c3e762251cbcf988f6962199a956 (MD5) Previous issue date: 2017-07-17 / Este trabalho investiga a utilização das metodologias ativas como instrumento para o ensino de Física a partir do uso de um “Kit Robótico” desenvolvido com uma plataforma de hardware livre chamada Arduíno. O planejamento do trabalho teve início em 2015, momento em que foi realizada uma revisão da literatura sobre a temática e sobre as metodologias que poderiam a ser adotadas no seu desenvolvimento. Como resultado da revisão teórico-metodológica optamos por implementar o projeto por meio de um conjunto de atividades baseadas na metodologia de Aprendizagem Baseada em Problemas focalizando a temática “Movimento”, dentro do currículo de Física do 1° ano do Ensino Médio de uma escola Técnica do Centro Estadual de Educação Tecnológica Paula Souza. O estudo de campo foi desenvolvido a partir de encontros periódicos com um grupo de alunos que manifestaram interesse pela proposta. Para direcionar as atividades e garantir que os alunos desenvolvessem as competências e habilidades desejadas foi utilizado os denominados os Roteiros de Atividades, que criava situações problemas para contextualizar a aprendizagem. Após a realização da pesquisa, verificamos alguns indícios de que o recursos e metodologias adotados podem auxiliar o professor e contribuir para o processo de ensino e aprendizagem, entretanto, não é a inserção da tecnologia, por si, o elemento fundamental que altera as condições de aprendizagem e sim de saber aproveitar as características que esses recursos possuem para fazer o aluno acreditar na legitimidade dos conteúdos escolares e envolver-se na construção do seu conhecimento. Deste modo, podemos dizer que, a ferramenta tecnológica utilizada nesta pesquisa (Kit Robótico), permitiu desenvolver satisfatoriamente atividades relativas ao tema “Movimento”, que atuou como elemento contextualizador e motivador no processo de aprendizagem. / This work investigates the use of active methodologies as an instrument for teaching physics from the use of a "Robotic Kit" developed with a free hardware platform called Arduino. Work planning began in 2015, at which time a review of the literature on the subject and on the methodologies that could be adopted in its development was carried out. As a result of the theoretical-methodological revision we decided to implement the project through a set of activities based on the methodology of Problem-Based Learning focusing on the theme "Movement" within the Physics curriculum of the 1st year of High School of a Technical School Of the State Center of Technological Education Paula Souza. The field study was developed from periodic meetings with a group of students who expressed an interest in the proposal. In order to direct the activities and ensure that the students developed the desired skills and abilities, we used the so-called Activity Schedules, which created situations to contextualize the learning situations. After conducting the research, we verified some evidence that the resources and methodologies adopted can help the teacher and contribute to the teaching and learning process, however, it is not the insertion of technology, by itself, the fundamental element that changes the conditions of Learning but rather to know how to take advantage of the characteristics that these resources have in order to make the student believe in the legitimacy of the school contents and to be involved in the construction of their knowledge. In this way, we can say that, the technological tool used in this research (Robotic Kit), allowed to develop satisfactorily activities related to the theme "Movement", which acted as contextualizing and motivating element in the learning process.

abstract: The rapid escalation of technology and the widespread emergence of modern technological equipments have resulted in the generation of humongous amounts of digital data (in the form of images, videos and text). This has expanded the possibility of solving real world problems using computational learning frameworks. However, while gathering a large amount of data is cheap and easy, annotating them with class labels is an expensive process in terms of time, labor and human expertise. This has paved the way for research in the field of active learning. Such algorithms automatically select the salient and exemplar instances from large quantities of unlabeled data and are effective in reducing human labeling effort in inducing classification models. To utilize the possible presence of multiple labeling agents, there have been attempts towards a batch mode form of active learning, where a batch of data instances is selected simultaneously for manual annotation. This dissertation is aimed at the development of novel batch mode active learning algorithms to reduce manual effort in training classification models in real world multimedia pattern recognition applications. Four major contributions are proposed in this work: $(i)$ a framework for dynamic batch mode active learning, where the batch size and the specific data instances to be queried are selected adaptively through a single formulation, based on the complexity of the data stream in question, $(ii)$ a batch mode active learning strategy for fuzzy label classification problems, where there is an inherent imprecision and vagueness in the class label definitions, $(iii)$ batch mode active learning algorithms based on convex relaxations of an NP-hard integer quadratic programming (IQP) problem, with guaranteed bounds on the solution quality and $(iv)$ an active matrix completion algorithm and its application to solve several variants of the active learning problem (transductive active learning, multi-label active learning, active feature acquisition and active learning for regression). These contributions are validated on the face recognition and facial expression recognition problems (which are commonly encountered in real world applications like robotics, security and assistive technology for the blind and the visually impaired) and also on collaborative filtering applications like movie recommendation. / Dissertation/Thesis / Ph.D. Computer Science 2013