Semi-supervised subspace clustering and applications to neuroscience

Machine learning techniques are used for extracting valuable knowledge from data. Nowa¬days, these techniques are becoming even more important due to the evolution in data ac¬quisition and storage, which is leading to data with different characteristics that must be exploited. Therefore, advances in data collection must be accompanied with advances in machine learning techniques to solve new challenges that might arise, on both academic and real applications. There are several machine learning techniques depending on both data characteristics and purpose. Unsupervised classification or clustering is one of the most known techniques when data lack of supervision (unlabeled data) and the aim is to discover data groups (clusters) according to their similarity. On the other hand, supervised classification needs data with supervision (labeled data) and its aim is to make predictions about labels of new data. The presence of data labels is a very important characteristic that guides not only the learning task but also other related tasks such as validation. When only some of the available data are labeled whereas the others remain unlabeled (partially labeled data), neither clustering nor supervised classification can be used. This scenario, which is becoming common nowadays because of labeling process ignorance or cost, is tackled with semi-supervised learning techniques. This thesis focuses on the branch of semi-supervised learning closest to clustering, i.e., to discover clusters using available labels as support to guide and improve the clustering process. Another important data characteristic, different from the presence of data labels, is the relevance or not of data features. Data are characterized by features, but it is possible that not all of them are relevant, or equally relevant, for the learning process. A recent clustering tendency, related to data relevance and called subspace clustering, claims that different clusters might be described by different feature subsets. This differs from traditional solutions to data relevance problem, where a single feature subset (usually the complete set of original features) is found and used to perform the clustering process. The proximity of this work to clustering leads to the first goal of this thesis. As commented above, clustering validation is a difficult task due to the absence of data labels. Although there are many indices that can be used to assess the quality of clustering solutions, these validations depend on clustering algorithms and data characteristics. Hence, in the first goal three known clustering algorithms are used to cluster data with outliers and noise, to critically study how some of the most known validation indices behave. The main goal of this work is however to combine semi-supervised clustering with subspace clustering to obtain clustering solutions that can be correctly validated by using either known indices or expert opinions. Two different algorithms are proposed from different points of view to discover clusters characterized by different subspaces. For the first algorithm, available data labels are used for searching for subspaces firstly, before searching for clusters. This algorithm assigns each instance to only one cluster (hard clustering) and is based on mapping known labels to subspaces using supervised classification techniques. Subspaces are then used to find clusters using traditional clustering techniques. The second algorithm uses available data labels to search for subspaces and clusters at the same time in an iterative process. This algorithm assigns each instance to each cluster based on a membership probability (soft clustering) and is based on integrating known labels and the search for subspaces into a model-based clustering approach. The different proposals are tested using different real and synthetic databases, and comparisons to other methods are also included when appropriate. Finally, as an example of real and current application, different machine learning tech¬niques, including one of the proposals of this work (the most sophisticated one) are applied to a task of one of the most challenging biological problems nowadays, the human brain model¬ing. Specifically, expert neuroscientists do not agree with a neuron classification for the brain cortex, which makes impossible not only any modeling attempt but also the day-to-day work without a common way to name neurons. Therefore, machine learning techniques may help to get an accepted solution to this problem, which can be an important milestone for future research in neuroscience. Resumen Las técnicas de aprendizaje automático se usan para extraer información valiosa de datos. Hoy en día, la importancia de estas técnicas está siendo incluso mayor, debido a que la evolución en la adquisición y almacenamiento de datos está llevando a datos con diferentes características que deben ser explotadas. Por lo tanto, los avances en la recolección de datos deben ir ligados a avances en las técnicas de aprendizaje automático para resolver nuevos retos que pueden aparecer, tanto en aplicaciones académicas como reales. Existen varias técnicas de aprendizaje automático dependiendo de las características de los datos y del propósito. La clasificación no supervisada o clustering es una de las técnicas más conocidas cuando los datos carecen de supervisión (datos sin etiqueta), siendo el objetivo descubrir nuevos grupos (agrupaciones) dependiendo de la similitud de los datos. Por otra parte, la clasificación supervisada necesita datos con supervisión (datos etiquetados) y su objetivo es realizar predicciones sobre las etiquetas de nuevos datos. La presencia de las etiquetas es una característica muy importante que guía no solo el aprendizaje sino también otras tareas relacionadas como la validación. Cuando solo algunos de los datos disponibles están etiquetados, mientras que el resto permanece sin etiqueta (datos parcialmente etiquetados), ni el clustering ni la clasificación supervisada se pueden utilizar. Este escenario, que está llegando a ser común hoy en día debido a la ignorancia o el coste del proceso de etiquetado, es abordado utilizando técnicas de aprendizaje semi-supervisadas. Esta tesis trata la rama del aprendizaje semi-supervisado más cercana al clustering, es decir, descubrir agrupaciones utilizando las etiquetas disponibles como apoyo para guiar y mejorar el proceso de clustering. Otra característica importante de los datos, distinta de la presencia de etiquetas, es la relevancia o no de los atributos de los datos. Los datos se caracterizan por atributos, pero es posible que no todos ellos sean relevantes, o igualmente relevantes, para el proceso de aprendizaje. Una tendencia reciente en clustering, relacionada con la relevancia de los datos y llamada clustering en subespacios, afirma que agrupaciones diferentes pueden estar descritas por subconjuntos de atributos diferentes. Esto difiere de las soluciones tradicionales para el problema de la relevancia de los datos, en las que se busca un único subconjunto de atributos (normalmente el conjunto original de atributos) y se utiliza para realizar el proceso de clustering. La cercanía de este trabajo con el clustering lleva al primer objetivo de la tesis. Como se ha comentado previamente, la validación en clustering es una tarea difícil debido a la ausencia de etiquetas. Aunque existen muchos índices que pueden usarse para evaluar la calidad de las soluciones de clustering, estas validaciones dependen de los algoritmos de clustering utilizados y de las características de los datos. Por lo tanto, en el primer objetivo tres conocidos algoritmos se usan para agrupar datos con valores atípicos y ruido para estudiar de forma crítica cómo se comportan algunos de los índices de validación más conocidos. El objetivo principal de este trabajo sin embargo es combinar clustering semi-supervisado con clustering en subespacios para obtener soluciones de clustering que puedan ser validadas de forma correcta utilizando índices conocidos u opiniones expertas. Se proponen dos algoritmos desde dos puntos de vista diferentes para descubrir agrupaciones caracterizadas por diferentes subespacios. Para el primer algoritmo, las etiquetas disponibles se usan para bus¬car en primer lugar los subespacios antes de buscar las agrupaciones. Este algoritmo asigna cada instancia a un único cluster (hard clustering) y se basa en mapear las etiquetas cono-cidas a subespacios utilizando técnicas de clasificación supervisada. El segundo algoritmo utiliza las etiquetas disponibles para buscar de forma simultánea los subespacios y las agru¬paciones en un proceso iterativo. Este algoritmo asigna cada instancia a cada cluster con una probabilidad de pertenencia (soft clustering) y se basa en integrar las etiquetas conocidas y la búsqueda en subespacios dentro de clustering basado en modelos. Las propuestas son probadas utilizando diferentes bases de datos reales y sintéticas, incluyendo comparaciones con otros métodos cuando resulten apropiadas. Finalmente, a modo de ejemplo de una aplicación real y actual, se aplican diferentes técnicas de aprendizaje automático, incluyendo una de las propuestas de este trabajo (la más sofisticada) a una tarea de uno de los problemas biológicos más desafiantes hoy en día, el modelado del cerebro humano. Específicamente, expertos neurocientíficos no se ponen de acuerdo en una clasificación de neuronas para la corteza cerebral, lo que imposibilita no sólo cualquier intento de modelado sino también el trabajo del día a día al no tener una forma estándar de llamar a las neuronas. Por lo tanto, las técnicas de aprendizaje automático pueden ayudar a conseguir una solución aceptada para este problema, lo cual puede ser un importante hito para investigaciones futuras en neurociencia.

Mejora de la enseñanza de la Geología mediante e-learning

Resumen: en este trabajo se presentan nuevas estrategias de enseñanza y aprendizaje a través de las nuevas tecnologías en su variante virtual a distancia (e-learning) implementadas en asignaturas relacionadas con la Geología. El objetivo básico fue acercar los aspectos geológicos a los estudiantes mediante el empleo de estas tecnologías. Se ha observado una mayor motivación y adquisición de conocimientos geológicos por parte del alumnado, que se ha traducido en una mejora en las calificaciones. Abstract: This paper deals with new teaching and learning approaches through the use of new technologies, mainly virtual distance learning (e-learning) in courses related to Geology. The main objective is to bring the geological aspects of Nature to students using these technologies. These new approaches have produced an increase in student motivation and acquisition of geological knowledge, accompanied by an improvement in their grades.

Web-based education in Spanish Universities. A Comparison of Open Source E-Learning Platforms

Web-based education or „e-learning‟ has become a critical component in higher education for the last decade, replacing other distance learning methods, such as traditional computer training or correspondence learning. The number of university students who take on-line courses is continuously increasing all over the world. In Spain, nearly a 90% of the universities have an institutional e-learning platform and over 60% of the traditional on-site courses use this technology as a supplement to the traditional face-to-face classes. This new form of learning allows the disappearance of geographical barriers and enables students to schedule their own learning process, among some other advantages. On-line education is developed through specific software called „e-learning platform‟ or „virtual learning environment‟ (VLE). A considerable number of web-based tools to deliver distance courses are currently available. Open source software packages such as Moodle, Sakai, dotLRN or Dokeos are the most commonly used in the virtual campuses of Spanish universities. This paper analyzes the possibilities that virtual learning environments provide university teachers and learners and offers a technical comparison among some of the most popular e-learning learning platforms.

c&d-learning implementation architecture: Adapting e-learning to developing countries

From the educational point of view, the most widespread method in developing countries is on-site education. Technical and economic resources cannot support conventional distance learning infrastructures and it is even worse for courses in universities. They usually suffer a lack of qualified faculty staff, especially in technical degrees. The literature suggest that e-learning is a suitable solution for this problem, but its methods are developed attending to educational necessities of the First World and cannot be applied directly to other contexts. The proposed methodology is a variant of traditional e-learning adapted to the needs of developing countries. E-learning for Cooperation and Development (c&d-learning) is oriented to be used for educational institutions without adequate technical or human resources. In this paper we describe the c&d-learning implementation architecture based on three main phases: hardware, communication and software; e.g. computer and technical equipping, internet accessing and e-learning platform adaptation. Proper adaptation of educational contents to c&d-learning is discussed and a real case of application in which the authors are involved is described: the Ngozi University at Burundi.

SONIA REGINA PRADO

A dissertação teve como objetivo principal estudar como uma Instituição de Ensino Superior Privada (IES) atuante no Brasil tem crescido pós Lei de Diretrizes e Bases (LDB) de 1996 até 2015, por meio da análise do curso de bacharelado em Administração de Empresas, nas modalidades: presencial, EAD e Flex (semipresencial). Para este fim, foi realizada uma pesquisa exploratória, de caráter qualitativo baseada no método do estudo de caso. Para coleta de evidências foram analisados relatórios corporativos (Annual Report, Relatórios Internos e outros documentos), entrevistas baseadas em roteiro semiestruturado com gestores da IES privada e observações. Dentre os principais achados, verificou-se que as principais estratégicas de crescimento da IES privada estudada se basearam em fusões e aquisições de outras IES, abertura de novos polos de EAD, na abertura de novas unidades próprias, bem como em inovações em várias dimensões da organização. Os programas governamentais de financiamento aos alunos também são fortes contribuintes para este crescimento, como o Fundo de Financiamento ao Estudante do Ensino Superior (FIES) e o Programa Universidade para Todos (Prouni). Com essa nova realidade, o ensino superior privado recebeu incentivo e facilitação para o seu crescimento, a um ritmo acelerado. Consequentemente pode-se concluir que a IES privada estudada adotou as seguintes estratégias de crescimento: Expansão orgânica com fusões/ aquisições de Instituições menores, com desenvolvimento de planos para todos os campi Brasil; Greenfield (por meio de solicitação de autorização de novas unidades e/ou cursos) em cidades sem possibilidades de aquisições/fusões, e aumentando o número de vagas/ matriculas nas unidades já existentes, aderiu aos programas do governo e também cuidou da evasão por meio de: Seguro educacional; gestão preparada para atender necessidades do discente; Sistema de Ensino com currículos integrados nacionalmente; Intercâmbio de alunos e professores entre as diversas unidades em todas as regiões do país e padronização dos processos.

Investigating and Understanding Student Learning Outcomes in an Online and Face-to-Face Graduate Level Legal Administration Course: An Embedded Mixed Methods Design

Online education is a new teaching and learning medium with few current guidelines for faculty, administrators or students. Its rapid growth over the last decade has challenged academic institutions to keep up with the demand, while also providing a quality education. Our understanding of the factors that determine quality and effective online learning experiences that lead to student learning outcomes is still evolving. There is a lack of consensus on the effectiveness of online versus face-to-face education in the current research. The U.S. Department of Education conducted a meta-analysis in 2009 and concluded that student-learning outcomes in online courses were equal to and, often times, better than face-to-face traditional courses. Subsequent research has found contradictory findings, and further inquiry is necessary. The purpose of this embedded mixed methods design research study is to further our understanding of the factors that create quality and successful educational outcomes in an online course. To achieve this, the first phase of this study measured and compared learning outcomes in an online and in class graduate-level legal administration course. The second phase of the study entailed interviews with those students in both the online and face-to-face sections to understand their perspectives on the factors contributing to learning outcomes. Six themes emerged from the qualitative findings: convenience, higher order thinking, discussions, professor engagement, professor and student interaction, and face-to-face interaction. Findings from this study indicate the factors students perceive as contributing to learning outcomes in an online course are consistent among all students and are supported in the existing literature. Higher order thinking, however, emerged as a stronger theme than indicated in the current research, and the face-to-face nature of the traditional classroom may be more an issue of familiarity than a factor contributing to learning outcomes. As education continues to reach new heights and developments in technology advance, the factors found to contribute to student learning outcomes will be refined and enhanced. These developments will continue to transform the ways in which we deliver and receive knowledge in both traditional and online classrooms. While there is a growing body of research on online education, the field’s evolution has unsettled earlier findings and posed new areas to investigate.

Problem-based learning: a case study in providing e-health education using the Internet

The current trend among many universities is to increase the number of courses available online. However, there are fundamental problems in transferring traditional education courses to virtual formats. Delivering current curricula in an online format does not assist in overcoming the negative effects on student motivation which are inherent in providing information passively. Using problem-based learning (PBL) online is a method by which computers can become a tool to encourage active learning among students. The delivery of curricula via goal-based scenarios allows students to learn at different rates and can successfully shift online learning from memorization to discovery. This paper reports on a Web-based e-health course that has been delivered via PBL for the past 12 months. Thirty distance-learning students undertook postgraduate courses in e-health delivered via the Internet (asynchronous communication). Data collected via online student surveys indicated that the PBL format was both flexible and interesting. PBL has the potential to increase the quality of the educational experience of students in online environments.

The judicious utilization of new technologies through authentic learning in higher education: A case study

Though technology holds significant promise for enhanced teaching and learning it is unlikely to meet this promise without a principled approach to course design. There is burgeoning discourse about the use of technological tools and models in higher education, but much of the discussion is fixed upon distance learning or technology based courses. This paper will develop and propose a balanced model for effective teaching and learning for “on campus” higher education, with particular emphasis on the opportunities for revitalisation available through the judicious utilisation of new technologies. It will explore the opportunities available for the creation of more authentic learning environments through the principled design. Finally it will demonstrate with a case study how these have come together enabling the creation of an effective and authentic learning environment for one pre-service teacher education course at the University of Queensland.

Vínculos afetivos na Educação a Distância: possibilidades e impossibilidades

Esta pesquisa analisou como estão sendo estabelecidos os vínculos afetivos nas relações de ensino-aprendizagem na modalidade de educação a distância. O trabalho foi desenvolvido dentro da perspectiva sócio-histórico-cultural que considera as emoções como constitutivas do pensamento, participantes do processo de significação e produção de sentidos. Durante a realização da pesquisa procuramos compreender a trajetória da EAD no Brasil, bem como verificar como está essa modalidade educativa hoje. Buscamos, ainda, explicitar as implicações da afetividade sobre o processo de ensino-aprendizagem, investigando manifestações de afetividade na modalidade de educação a distância. Para isso, nos baseamos em alguns autores, em especial, na teoria de Henri Wallon sobre o desenvolvimento humano. Para alcançar os objetivos deste trabalho fizemos uma retomada do processo histórico da EAD no Brasil, refletindo sobre a formação dos educadores e como essa modalidade tem sido desenvolvida no nosso país. Em um segundo momento, aprofundamos a discussão sobre afetividade e a criação de vínculos na EAD. Após essas etapas, partimos para a pesquisa de campo que consistiu em investigar 10 (dez) sujeitos envolvidos na modalidade: 2 (dois) professores temáticos, 2 (dois) professores-tutores e 6 (seis) alunos que estudam na modalidade de três instituições diferentes, sendo uma pública e duas privadas. Utilizamos como instrumentos de pesquisa questionários e roteiros semi-estruturados de entrevistas para aprofundar algumas questões. As análises demonstram que os vínculos afetivos entre professor-aluno são primordiais para a aprendizagem, definindo-se como condição imprescindível para o desenvolvimento cognitivo dentro do espaço escolar e na sociedade, e estes são estabelecidos tanto na modalidade presencial quanto na modalidade a distância, sendo fundamentais para que ocorra uma aprendizagem significativa.

ESTRATÉGIAS DE EXPANSÃO DE INSTITUIÇÕES DE ENSINO SUPERIOR PRIVADAS NO BRASIL: ESTUDO DE CASO DO CURSO DE BACHARELADO EM ADMINISTRAÇÃO

A dissertação teve como objetivo principal estudar como uma Instituição de Ensino Superior Privada (IES) atuante no Brasil tem crescido pós Lei de Diretrizes e Bases (LDB) de 1996 até 2015, por meio da análise do curso de bacharelado em Administração de Empresas, nas modalidades: presencial, EAD e Flex (semipresencial). Para este fim, foi realizada uma pesquisa exploratória, de caráter qualitativo baseada no método do estudo de caso. Para coleta de evidências foram analisados relatórios corporativos (Annual Report, Relatórios Internos e outros documentos), entrevistas baseadas em roteiro semiestruturado com gestores da IES privada e observações. Dentre os principais achados, verificou-se que as principais estratégicas de crescimento da IES privada estudada se basearam em fusões e aquisições de outras IES, abertura de novos polos de EAD, na abertura de novas unidades próprias, bem como em inovações em várias dimensões da organização. Os programas governamentais de financiamento aos alunos também são fortes contribuintes para este crescimento, como o Fundo de Financiamento ao Estudante do Ensino Superior (FIES) e o Programa Universidade para Todos (Prouni). Com essa nova realidade, o ensino superior privado recebeu incentivo e facilitação para o seu crescimento, a um ritmo acelerado. Consequentemente pode-se concluir que a IES privada estudada adotou as seguintes estratégias de crescimento: Expansão orgânica com fusões/ aquisições de Instituições menores, com desenvolvimento de planos para todos os campi Brasil; Greenfield (por meio de solicitação de autorização de novas unidades e/ou cursos) em cidades sem possibilidades de aquisições/fusões, e aumentando o número de vagas/ matriculas nas unidades já existentes, aderiu aos programas do governo e também cuidou da evasão por meio de: Seguro educacional; gestão preparada para atender necessidades do discente; Sistema de Ensino com currículos integrados nacionalmente; Intercâmbio de alunos e professores entre as diversas unidades em todas as regiões do país e padronização dos processos.

Protein-protein interactions classification from text via local learning with class priors

Text classification is essential for narrowing down the number of documents relevant to a particular topic for further pursual, especially when searching through large biomedical databases. Protein-protein interactions are an example of such a topic with databases being devoted specifically to them. This paper proposed a semi-supervised learning algorithm via local learning with class priors (LL-CP) for biomedical text classification where unlabeled data points are classified in a vector space based on their proximity to labeled nodes. The algorithm has been evaluated on a corpus of biomedical documents to identify abstracts containing information about protein-protein interactions with promising results. Experimental results show that LL-CP outperforms the traditional semisupervised learning algorithms such as SVMand it also performs better than local learning without incorporating class priors.

The importance of power dynamics in the development of asynchronous online learning communities

This research explored how a more student-directed learning design can support the creation of togetherness and belonging in a community of distance learners in formal higher education. Postgraduate students in a New Zealand School of Education experienced two different learning tasks as part of their online distance learning studies. The tasks centered around two online asynchronous discussions each for the same period of time and with the same group of students, but following two different learning design principles. All messages were analyzed using a twostep analysis process, content analysis and social network analysis. Although the findings showed a balance of power between the tutor and the students in the first high e-moderated activity, a better pattern of group interaction and community feeling was found in the low e-moderated activity. The paper will discuss the findings in terms of the implications for learning design and the role of the tutor.