教育部從1997年開始協助大專院校設立推廣教育機構，以有效利用大學資源對社會人士從事終身學習，協助設計相關專業課程。在市場激烈競爭下，大學在規劃推廣教育的策略方向時，需考量全球化、數位化、少子化的市場特性。 　　本研究嘗試以司徒達賢(1999)提出之非營利組織策略理論CORPS (Client, Operation, Resource, Participate and Service)為基礎 配合武文瑛(2006)的回流教育指標進行策略研究。本研究以某大學之推廣教育機構做為研究個案，針對學員與職員進行問卷調查與個別訪談，獲得以下幾點發現： 一、 學員是因為大學名聲與師資才重回校園學習，但費用高低與設備新舊會影響學習意願與成效。 二、 推廣教育應增加創新思維的課程並且提升教學品質，在追求課程數位化之餘仍應重視課程內容。 三、 整合大學內部資源才有機會提升推廣教育的成效，爭取各系所師資與強化作業流程，才能提供學員更優質的服務。 / The Ministry of Education helped universities and colleges to setup extension education centers from 1997. The Ministry of Education urged the universities to design programs for professionals by use of their own resource. Universities build up a lot of classes with credits and degrees in recent years. The competition between universities makes them focusing on the strategies confronting different aspects such as globalization, digitization and low birth rate. 　　This research focuses on strategy analysis and refers to nonprofit organization theory proposed by Seetoo(1999)－CORPS (Client, Operation, Resource, Participate and Service) and recurrent education indicators suggested by Wu(2006). Based on survey and interview with students and employees at a certain extension education center, this study has findings as follows: 1. The reason for professionals returning to extension education center is because the university reputation and well-known professors. The tuition and facilities affect their willingness to learn and the effect of learning. 2. The organization should provide innovation and high quality courses not just investing on digital technology. 3. The integration of university’s resource will increase the benefits of extension education. It will provide better services with process improvement and cooperation between different academic departments.

In this paper a novel speaker recognition system is introduced. Automated speaker recognition has become increasingly popular to aid in crime investigations and authorization processes with the advances in computer science. Here, a recurrent neural network approach is used to learn to identify ten speakers within a set of 21 audio books. Audio signals are processed via spectral analysis into Mel Frequency Cepstral Coefficients that serve as speaker specific features, which are input to the neural network. The Long Short-Term Memory algorithm is examined for the first time within this area, with interesting results. Experiments are made as to find the optimum network model for the problem. These show that the network learns to identify the speakers well, text-independently, when the recording situation is the same. However the system has problems to recognize speakers from different recordings, which is probably due to noise sensitivity of the speech processing algorithm in use.

Inferring Genetic Regulatory Networks (GRN) from multiple data sources is a fundamental problem in computational biology. Computational models for GRN range from simple Boolean networks to stochastic differential equations. To successfully model GRN, a computational method has to be scalable and capable of integrating different biological data sources effectively and homogeneously. In this thesis, we introduce a novel method to model GRN using Cost-Based Abduction (CBA) and study the relation between CBA and Bayesian inference. CBA is an important AI formalism for reasoning under uncertainty that can integrate different biological data sources effectively. We use three different yeast genome data sources—protein-DNA, protein-protein, and knock-out data—to build a skeleton (unannotated) graph which acts as a theory to build a CBA system. The Least Cost Proof (LCP) for the CBA system fully annotates the skeleton graph to represent the learned GRN. Our results show that CBA is a promising tool in computational biology in general and in GRN modeling in particular because CBA knowledge representation can intrinsically implement the AND/OR logic in GRN while enforcing cis-regulatory logic constraints effectively, allowing the method to operate on a genome-wide scale.Besides allowing us to successfully learn yeast pathways such as the pheromone pathway, our method is scalable enough to analyze the full yeast genome in a single CBA instance, without sub-networking. The scalability power of our method comes from the fact that our CBA model size grows in a quadratic, rather than exponential, manner with respect to data size and path length. We also introduce a new algorithm to convert CBA into an equivalent binary linear program that computes the exact LCP for the CBA system, thus reaching the optimal solution. Our work establishes a framework to solve Bayesian networks using integer linear programming and high order recurrent neural networks through CBA as an intermediate representation.

Incomplete data often arise in the study of life history processes. Examples include missing responses, missing covariates, and unobservable latent processes in addition to right censoring. This thesis is on the development of statistical models and methods to address these problems as they arise in oncology and chronic disease. Methods of estimation and inference in parametric, weakly parametric and semiparametric settings are investigated. Studies of chronic diseases routinely sample individuals subject to conditions on an event time of interest. In epidemiology, for example, prevalent cohort studies aiming to evaluate risk factors for survival following onset of dementia require subjects to have survived to the point of screening. In clinical trials designed to assess the effect of experimental cancer treatments on survival, patients are required to survive from the time of cancer diagnosis to recruitment. Such conditions yield samples featuring left-truncated event time distributions. Incomplete covariate data often arise in such settings, but standard methods do not deal with the fact that the covariate distribution is also affected by left truncation. We develop a likelihood and algorithm for estimation for dealing with incomplete covariate data in such settings. An expectation-maximization algorithm deals with the left truncation by using the covariate distribution conditional on the selection criterion. An extension to deal with sub-group analyses in clinical trials is described for the case in which the stratification variable is incompletely observed. In studies of affective disorder, individuals are often observed to experience recurrent symptomatic exacerbations of symptoms warranting hospitalization. Interest lies in modeling the occurrence of such exacerbations over time and identifying associated risk factors to better understand the disease process. In some patients, recurrent exacerbations are temporally clustered following disease onset, but cease to occur after a period of time. We develop a dynamic mover-stayer model in which a canonical binary variable associated with each event indicates whether the underlying disease has resolved. An individual whose disease process has not resolved will experience events following a standard point process model governed by a latent intensity. If and when the disease process resolves, the complete data intensity becomes zero and no further events will arise. An expectation-maximization algorithm is developed for parametric and semiparametric model fitting based on a discrete time dynamic mover-stayer model and a latent intensity-based model of the underlying point process. The method is applied to a motivating dataset from a cohort of individuals with affective disorder experiencing recurrent hospitalization for their mental health disorder. Interval-censored recurrent event data arise when the event of interest is not readily observed but the cumulative event count can be recorded at periodic assessment times. Extensions on model fitting techniques for the dynamic mover-stayer model are discussed and incorporate interval censoring. The likelihood and algorithm for estimation are developed for piecewise constant baseline rate functions and are shown to yield estimators with small empirical bias in simulation studies. Data on the cumulative number of damaged joints in patients with psoriatic arthritis are analysed to provide an illustrative application.

Like many industries, the nuclear power industry in Sweden is currently facing the challenges of a major generational change. To meet these challenges, alongside the demands for a high level of security, the industry has attempted to standardise its mode of operations as far as possible. Apart from various technological fixes and safety devices, manuals and instructions have been modelled for every conceivable situation, or course of events; documentations and formal systems of co-ordination that become larger and larger, and more and more detailed. In high-risk industries there is a tendency to equate learning with changes in external patterns of behaviour, as against fixed standards, typically among operating staff. The acquisition  of professional skill, on the other hand, is the result of participation in practice. From this point of view, rather, learning is the outcome of reflection, upon actual events and experiences. Recurrent training can be used to promote formalisation, but also to explore and reinforce the experience based knowledge of skilled operators; between these approaches, the former prevails. Accidents and incidents incessantly put in question what is commonly referred to as the safety culture of various power plants, and subsequent to the misfortunes at Forsmark 1 in 2006, the accident was described as the culmination of a longterm decline in safety culture. The strong requirement for security and control is a cause of formalisation, whereas the need to support reflection as formation of professional skill tends to be omitted. Even so, experience based skill and knowledge remains a substantial consituent of what could be regarded as a dependable safety culture. Codified knowledge must be interpreted and applied in practice. Furthermore, experienced professionals, from encountering a great variety of situations, seem to develop what can be described as the skill of anticipation, and, as shown in connection with the incident at Forsmark 1, an ability to handle the unexpected. The urge for formalisation raises certain concerns: that of the primacy of defining the containments of professional skill, the impact and resilience of local knowledge and diversity, and the hollowing out of ability and skill within work-life organisations. The “human factor”, that is the operating staff, is commonly made responsible for established accidents and incidents. Even so, experienced personnel are able to manage a variety of unforeseen events and disturbances, that sometimes occur in high-risk technology industries. At times, on the contrary, the human factor saves technology, instead of the other way around. This study explores the concept of safety culture within the nuclear power industry from an epistemological perspective. It discusses the use of recurrent training, and the role of experience based skill and knowledge in the operating of Swedish power plants. What methods can be employed to support experience based knowledge as an essential complement to standardised work processes, codified knowledge, or benchmark strategies? Principles of formalisation need to be supplemented with a more thorough exploration of professional skill, in which a distinction between behaviour and responsibility can be made. / QC 20110906

[Truncated abstract] Field pea (Pisum sativum) has many benefits when included in the crop rotation system in broadacre grain farming. These benefits include a disease break and improved weed control for cereals and less dependence on nitrogenous fertilisers due to the leguminous nature of pea. Currently, field pea adoption in Australia is low because the crop is susceptible to the fungal disease `black spot’ (Mycosphaerella pinodes) and has low stem strength and a lodged canopy. Black spot causes yield losses averaging 10-15% per year. Lodging results in difficult and costly harvesting, increased disease pressure and increased wind erosion from exposed soil surface when stems break at the basal nodes. This project aimed to address these problems through breeding, and through the application of quantitative genetics theory to a recurrent selection program. A quantitative measurement of relative stem strength was developed which could be used effectively in the field on single plants. Accurate laboratory measurements of stem strength were closely correlated with the field measure of compressed stem thickness in the basal node region. A diallel analysis of stem strength of the progeny of crosses among a range of pea lines with different values of compressed stem thickness concluded that the genetic control of stem strength was additive, with no maternal inheritance or dominance or epistasis effects.

Submitted by Oliveira Santos Dilzaná (dilznana@yahoo.com.br) on 2014-01-21T13:40:06Z No. of bitstreams: 1 VÍVIAN VOLKMER PONTES.pdf: 4564074 bytes, checksum: 60dcf7571f664deb09ea5711b8570d64 (MD5) / Approved for entry into archive by Ana Portela (anapoli@ufba.br) on 2014-02-03T14:22:59Z (GMT) No. of bitstreams: 1 VÍVIAN VOLKMER PONTES.pdf: 4564074 bytes, checksum: 60dcf7571f664deb09ea5711b8570d64 (MD5) / Made available in DSpace on 2014-02-03T14:22:59Z (GMT). No. of bitstreams: 1 VÍVIAN VOLKMER PONTES.pdf: 4564074 bytes, checksum: 60dcf7571f664deb09ea5711b8570d64 (MD5) / A experiência de uma perda gestacional involuntária representa uma ruptura da gestação em desenvolvimento, daquilo que se esperava e que estava na iminência de acontecer: o tornar-se mãe e o nascimento de um bebê. Deste modo, implica em descontinuidade do desenvolvimento do self, desafiando entendimentos e expectativas que haviam sido construídos acerca de si mesmo e do mundo. A repetição dessas rupturas ao longo da trajetória reprodutiva ameaça o sentido do self, intensifica o nível de ambivalência e maximiza a incerteza em relação ao futuro. O presente estudo consistiu em um aprofundamento teórico e empírico sobre como o self constrói continuidade através de rupturas sucessivas ao longo da trajetória reprodutiva. Implicou entender as dinâmicas do self dialógico no fluxo do tempo, e analisá-las com o intuito de identificar processos de reorganização do self a fim de se manter coeso – apesar das rupturas, das recorrentes descontinuidades. Tendo em vista esta problemática, optou-se por tratá-la a partir do referencial teórico do Construtivismo Semiótico-Cultural, uma perspectiva teórico-metodológica que focaliza a construção inter e intrapessoal da subjetividade humana individual. Como método, realizou-se um estudo de cunho etnográfico realizado em dois contextos distintos de assistência à saúde da mulher e/ou casal com diagnóstico de aborto de repetição, na cidade do Salvador/BA, sendo um vinculado à rede pública e o outro à rede privada. No que concerne às estratégias de coleta de dados, destaque especial foi concedido às entrevistas narrativas. Já com relação à análise dos dados, foi utilizado o Modelo de Equifinalidade de Trajetórias (TEM). Participaram deste estudo dez mulheres com história de perdas gestacionais recorrentes. Após a realização de comparações descritivas entre as trajetórias reprodutivas das participantes – com a finalidade de entender os processos de rupturas e reconstruções na cultura pessoal através da aplicação do TEM –, foi construído um conceito que pudesse oferecer generalidade para o específico dessa experiência. Afinal, a experiência recorrente de rupturas significativas, exige um tipo específico de processo semiótico, que no presente trabalho foi denominado de estratégias semióticas de reparação dinâmica do self. Essas estratégias, ao serem utilizadas, levam à construção de signos específicos – os signos reparadores –, que têm o poder de restaurar a conexão dos fragmentos da trajetória interrompida, construindo alguma articulação entre esses fragmentos e resgatando certo senso de continuidade. Os signos reparadores promovem, assim, a construção de significado para o momento presente, a reconstrução de significados atribuídos às experiências no passado (perdas anteriores) e nova orientação para a gama aceitável de construções de significados orientados para o futuro, conectando-os, relançando-os numa nova narrativa, sempre singular – e, de certo modo, unificada e coerente –, acerca de si mesmo, da sua própria vida e do seu próprio “destino”. The experience of an involuntary pregnancy loss represents a rupture related to what was expected and what was about to happen: becoming a mother and the birth of a baby. Thus, this event implies discontinuity the development of self, challenging understandings and expectations that had been built about the person and the world. The repetition of these ruptures along the trajectory reproductive threatens the sense of self, intensifies the level of ambivalence and maximizes uncertainty about the future. This study consisted in a theoretical and empirical deepening about how the self builds continuity through successive ruptures along the reproductive trajectory. It was set up understand the dynamics of the dialogical self in the flow of time, and analyze them in order to identify processes of reorganization of the self in order to remain cohesive - despite the ruptures, the recurrent discontinuities. In view of this problem, we chose to approach it from the theoretical framework of Semiotic-Cultural Constructivism, a theoretical-methodological framework that focuses on building inter and intrapersonal individual human subjectivity. As a method, the ethnographic study was conducted in two distinct contexts of health care for women and / or couples with a diagnosis of recurrent abortion, in Salvador / BA: one linked to the public network and another network private. Concerning data collection strategies, special attention was given to the narrative interviews. Regarding data analysis, we used the Trajectories Equifinality Model (TEM). The participants of study were ten women with a history of recurrent pregnancy loss. After conducting descriptive comparisons between reproductive trajectories of participants - in order to understand the processes of disruptions and reconstruction in personal culture through the application of TEM - was built a concept that could provide for the generality of this specific experience. After all, the recurring experience of significant disruptions requires a specific type of semiotic process, which in this work was named Semiotic Strategies of Dynamic Self-Repairing. These strategies, when used, lead to the construction of specific signs – repairing signs – who have the power to restore the connection of the fragments of the interrupted trajectory, building a link between these fragments and rescuing some sense of continuity. Repairing signs thereby promotes the construction of meaning for the present moment, the reconstruction of meanings assigned to experiences in the past (previous losses) and new guidance for the acceptable range of constructions of meaning future-oriented, connecting them, relaunching them into a new narrative, always singular - and somehow unified and consistent - about yourself, your own life and your own "destiny".

Název práce: Geometrické lineární a nelineární problémy prostor· funkcí Autor: Petr Petráček Katedra: Katedra matematické analýzy 'kolitel: prof. RNDr. Jaroslav Lukeš, DrSc., Katedra matematické analýzy Abstrakt: Tato práce sestává ze čtyř vědeckých článk·. lánky prezentované v prvních dvou kapitolách se věnují teorii reálných a komplexních L1-preduál·. lánky prezentované v třetí a čtvrté kapitole jsou věnovány problematice line- ability a algebrability podmnožin reálných funkcí a měr. V Kapitole 1 předsta- vujeme charakterizaci komplexních L1-preduál· pomocí komplexního barycent- rického zobrazení. Tato charakterizace je přirozeným rozšířením charakterizace reálných L1 preduál· pocházející od Bednara a Laceyho. V Kapitole 2 odpoví- dáme na otázku položenou Laceym v roce 1973. Dokazujeme přitom existenci kompaktního prostoru K a uzavřeného podprostoru H ⊂ C(K) obsahujícího kon- stantní funkce, pro který platí ∂HK = K, H je maximální vzhledem k ∂HK a H není L1-preduál. V Kapitole 3 se věnujeme lineabilitě množin nikde mono- tonních znaménkových Radonových měr na Rd . Konkrétně dokazujeme existence vektorového prostoru dimenze c jehož každý nenulový prvek je nikde monotonní míra absolutně spojitá vzhledem k d-rozměrné Lebesgueově míře. Nadto dokazu- jeme, že existuje takový lineární prostor, který je hustý...

Statistical shift-reduce parsing involves the interplay of representation learning, structured learning, and inexact search. This dissertation considers approaches that tightly integrate these three elements and explores three novel models for shift-reduce CCG parsing. First, I develop a dependency model, in which the selection of shift-reduce action sequences producing a dependency structure is treated as a hidden variable; the key components of the model are a dependency oracle and a learning algorithm that integrates the dependency oracle, the structured perceptron, and beam search. Second, I present expected F-measure training and show how to derive a globally normalized RNN model, in which beam search is naturally incorporated and used in conjunction with the objective to learn shift-reduce action sequences optimized for the final evaluation metric. Finally, I describe an LSTM model that is able to construct parser state representations incrementally by following the shift-reduce syntactic derivation process; I show expected F-measure training, which is agnostic to the underlying neural network, can be applied in this setting to obtain globally normalized greedy and beam-search LSTM shift-reduce parsers.

No description available.