551 |
"Putting your house in order" - an exploration of the idea of a good death among people dying in mid-lifeCharlton, Diana Eleanor Marjorie 16 November 2006 (has links)
Student Number : 7916069 -
MA research report -
School of Human and Community Development -
Faculty of Humanities / This qualitative study set out to explore the idea of a good death through
in-depth interviews with six terminally ill patients with cancer aged between
thirty-seven and fifty-two, in other words in mid-life. Thematic content analysis
was used to examine overt and covert themes revealed in semi-structured
interviews that had been transcribed verbatim. Significant fear of dying was
revealed and although respondents did not seem to have a conscious idea of a
good death, they had a clear concept of good dying. Two key components of
good dying were not being in pain or distress from physical symptoms, and
wanting to drift off into a final “sleep” rather than being fully alert until the end. It
was noticeable that, whether or not patients had religious convictions, relatively
little attention appeared to be paid directly to what might or might not follow the
moment of death itself, for example an afterlife. Preparation for good dying
included completing a will, sorting out financial affairs and, for some patients,
planning a funeral. At times this preparation also included trying to mend
conflicted relationships and make plans for the ongoing care of family
members. It is postulated that these preparations helped re-constitute a sense
of order that had been shattered by the chaos of being declared terminally ill at
a time of life when this was non-normative. Moreover, taking care of practical
needs re-established a sense of agency, helped achieve some sense of closure
and symbolised a measure of acceptance of their dying status. Respondents
did not seem to experience external pressure from others to die in a certain
way, although two people were particularly aware of their influence on how
others in the family felt and thus tended to pretend to feel better than they did.
|
552 |
An approach for online learning in the presence of concept changes / Une approche pour l'apprentissage en-ligne en présence de changements de concept.Jaber, Ghazal 18 October 2013 (has links)
De nombreuses applications de flux de données ont vu le jour au cours des dernières années. Lorsque l'environnement évolue, il est nécessaire de s'appuyer sur un apprentissage en ligne pouvant s'adapter aux conditions changeantes, alias dérives de concept. L'adaptation aux dérives de concept implique d'oublier une partie ou la totalité des connaissances acquises lorsque le concept change, tout en accumulant des connaissances sur le concept sous-jacent supposé stationnaire. Ce compromis est appelé le dilemme stabilité-plasticité.Les méthodes d'ensemble ont été parmi les approches les plus réussies. Cependant, la gestion de l'ensemble qui détermine les informations à oublier n'a pas été complètement étudiée jusqu'ici. Notre travail montre l'importance de la stratégie de l'oubli en comparant plusieurs approches. Les résultats ainsi obtenus nous amènent à proposer une nouvelle méthode d'ensemble avec une stratégie d'oubli conçue pour s'adapter aux dérives de concept. Des évaluations empiriques montrent que notre méthode se compare favorablement aux systèmes adaptatifs de l'état de l'art.Les majorité des anciens travaux réalisés se sont focalisés sur la détection des changements de concept, ainsi que les méthodes permettant d'adapter le système d'apprentissage aux changements. Dans ce travail, nous allons plus loin en introduisant un mécanisme d'anticipation capable de détecter des états pertinents de l'environnement, de reconnaître les contextes récurrents et d'anticiper les changements de concept susceptibles.Par conséquent, la méthode que nous proposons traite à la fois le défi d'optimiser le dilemme stabilité-plasticité, l'anticipation et la reconnaissance des futurs concepts. Ceci est accompli grâce à une méthode d'ensemble qui contrôle un comité d'apprenants. D'une part, la gestion de l'ensemble permet de s'adapter naturellement à la dynamique des changements de concept avec peu de paramètres à régler. D'autre part, un mécanisme d'apprentissage surveillant les changements dans l'ensemble fournit des moyens pour anticiper la modification sous-jacente du contexte. / Learning from data streams is emerging as an important application area. When the environment changes, it is necessary to rely on on-line learning with the capability to adapt to changing conditions a.k.a. concept drifts. Adapting to concept drifts entails forgetting some or all of the old acquired knowledge when the concept changes while accumulating knowledge regarding the supposedly stationary underlying concept. This tradeoff is called the stability-plasticity dilemma. Ensemble methods have been among the most successful approaches. However, the management of the ensemble which ultimately controls how past data is forgotten has not been thoroughly investigated so far. Our work shows the importance of the forgetting strategy by comparing several approaches. The results thus obtained lead us to propose a new ensemble method with an enhanced forgetting strategy to adapt to concept drifts. Experimental comparisons show that our method compares favorably with the well-known state-of-the-art systems. The majority of previous works focused only on means to detect changes and to adapt to them. In our work, we go one step further by introducing a meta-learning mechanism that is able to detect relevant states of the environment, to recognize recurring contexts and to anticipate likely concepts changes. Hence, the method we suggest, deals with both the challenge of optimizing the stability-plasticity dilemma and with the anticipation and recognition of incoming concepts. This is accomplished through an ensemble method that controls a ensemble of incremental learners. The management of the ensemble of learners enables one to naturally adapt to the dynamics of the concept changes with very few parameters to set, while a learning mechanism managing the changes in the ensemble provides means for the anticipation of, and the quick adaptation to, the underlying modification of the context.
|
553 |
Revisorns och klientens påverkan på revisionskvalitetEriksson, Tobias, Kullab, Ahmad January 2019 (has links)
Ett företags intressenter behöver kunna lägga stor tillit i företagets finansiella rapporter. Därför är det betydelsefullt att det finns en neutral part som revisorer för att säkerställa riktigheten i dessa finansiella rapporter. Lågkvalitativa revisioner kan leda till skadliga konsekvenser för intressenterna eftersom de inte kommer att kunna ta rätt beslut som baseras på dessa finansiella rapporter. Oegentligheter i reviderade årsredovisningar kan också få allvarliga konsekvenser för den ansvariga revisionsbyrån som riskerar ett skamfilat rykte, skadat varumärke och i längden minskade intäkter genom klientbortfall. Vikten av hög revisionskvalitet är grundläggande i företagsekonomin, men en stor faktor till osäkerhet kring begreppet revisionskvalitet finns i hur man mäter och definierar det. Denna studie undersöker hur olika variabler påverkar revisorers benägenhet att utfärda fortsatt drift-anmärkningar, där fortsatt drift-anmärkningar används som en proxy på revisionskvalitet. En fortsatt drift-anmärkning ska utfärdas av revisorn om denne tror att det finns risk för att ifrågasätta ett företags fortsatta drift. En revision som innehåller en fortsatt drift-anmärkning över ett företag som sedan går i konkurs inom tolv månader från revisionstillfället, anses vara av hög kvalitet. Revisionskvalitet har varit och är fortfarande en viktig fråga i flera tidigare studier om revision. Dessa studier har identifierat flera faktorer, både revisorsspecifika och klientspecifika, som kan påverka revisionskvaliteten. Till de faktorer som kan härledas till revisorn finns revisorns kön och revisionsbyråns storlek. Till faktorerna som kan härledas till klienten finns företagets storlek och olika finansiella mått på företagets ekonomiska ställning. Tidigare studier indikerar att dessa faktorer är viktiga för revisorns förmåga att utfärda korrekta revisioner rörande fortsatt drift. I denna studie ser vi närmare på dessa faktorer och deras inverkan på revisionskvaliteten i en svensk kontext. Studien genomfördes med en kvantitativ forskningsmetod och ett deduktivt synsätt där årsredovisningar och revisionsberättelser från 282 konkursinledda onoterade aktiebolag under perioden 2013–2017 undersöktes för att undersöka eventuella samband mellan studiens variabler och revisionskvalitet. Resultatet visar att revisorns kön, klientens soliditet och klientens kassalikviditet har en signifikant inverkan på utfärdandet av fortsatt drift-anmärkningar och därmed revisionskvaliteten. Vårt resultat ger dock inget stöd för påståendet att revisionsbyråns storlek, klientens storlek eller klientens räntabilitet påverkar revisionskvaliteten.
|
554 |
Transporte caótico causado por ondas de deriva / Chaotic Transport Driven by Drift WavesSuigh, Rafael Oliveira 07 December 2010 (has links)
Um dos problemas enfrentados pelos cientistas para o confinamento de plasma em Tokamaks, para se obter fusão termonuclear controlada, é o transporte radial de partículas pela borda do plasma. Nessa dissertação, estudamos o transporte através de um modelo que relaciona as flutuações eletrostáticas na borda do plasma às ondas de deriva. Essas ondas criam no plasma regiões de fluxo convectivo, formando ilhas que são, eventualmente, separadas por barreiras. Para apenas uma onda, o sistema é integrável e todas as trajetórias do plano de fase são curvas invariantes que, se não existirem barreiras, estão em ilhas divididas por separatrizes. Foi verificado que, quando uma segunda onda com velocidade de fase diferente da primeira é utilizada, o sistema não é mais integrável e a região anteriormente ocupada pelas separatrizes torna-se caótica. Com a quebra de separatrizes ocorre o transporte caótico de partículas. Quando uma separatriz é quebrada, surge em seu lugar uma estrutura que ainda preserva algumas características da separatriz, mas se modifica no espaço de fases ao longo do tempo. Essa estrutura é conhecida como Estrutura Lagrangiana Coerente (ELC). Nessa dissertação verificamos que as ELCs, por um lado, funcionam como barreiras de transporte, pois nenhuma trajetória consegue atravessa-la e, por outro lado, criam regiões no espaço de fases onde o transporte é alto, pois trajetórias próximas a elas tendem a ser aceleradas. Uma das principais contribuições obtidas ao se estudar ELCs no problema de duas ondas de deriva, aplicado ao confinamento de plasmas em Tokamaks, é a possibilidade de se prever a existência de ilhas, que funcionem como barreiras de transporte, no plano de fases que, por sua vez, são um importante mecanismo de aprisionamento de partículas. / One of the problems facing scientists in the confinement of plasma in tokamaks, to obtain controlled thermonuclear fusion, is the radial transport of particles at the plasma edge. In this dissertation, we study particle transport through a model that relates the electrostatic fluctuations at the edge of the plasma with drift waves. These waves create regions inside the plasma with convective flow, forming islands that are eventually separated by barriers. For one wave, the system is integrable and all the trajectories of phase space are invariant curves that are divided by separatrices. It was found that when a second wave with phase velocity different from the first is used, the system is no longer integrable and the region previously occupied by the separatrix becomes chaotic. With the destruction of the separatrix the transport of particles is chaotic. When a separatrix is broken, appears in its place a structure that preserves some features of the separatrix, but it is changing in phase space over time. This structure is known as Lagrangian Coherent Structure (LCS). In this dissertation we found that the LCSs, on the one hand, act as transport barriers, since no trajectory can cross it and, moreover, creates regions in phase space where particle transport is high, because trajectories close to them tend to be accelerated. One of the main contributions obtained by studying LCSs in the problem of two drift waves, applied to the confinement of plasma in tokamaks, is the ability to predict the existence of islands, which act as transport barriers, which are an important mechanism of trapping particles.
|
555 |
Classificação de fluxos de dados com mudança de conceito e latência de verificação / Data stream classification with concept drift and verification latencyReis, Denis Moreira dos 27 September 2016 (has links)
Apesar do grau relativamente alto de maturidade existente na área de pesquisa de aprendizado supervisionado em lote, na qual são utilizados dados originários de problemas estacionários, muitas aplicações reais lidam com fluxos de dados cujas distribuições de probabilidade se alteram com o tempo, ocasionando mudanças de conceito. Diversas pesquisas vêm sendo realizadas nos últimos anos com o objetivo de criar modelos precisos mesmo na presença de mudanças de conceito. A maioria delas, no entanto, assume que tão logo um evento seja classificado pelo algoritmo de aprendizado, seu rótulo verdadeiro se torna conhecido. Este trabalho explora as situações complementares, com revisão dos trabalhos mais importantes publicados e análise do impacto de atraso na disponibilidade dos rótulos verdadeiros ou sua não disponibilização. Ainda, propõe um novo algoritmo que reduz drasticamente a complexidade de aplicação do teste de hipótese não-paramétrico Kolmogorov-Smirnov, tornado eficiente seu uso em algoritmos que analisem fluxos de dados. A exemplo, mostramos sua potencial aplicação em um método de detecção de mudança de conceito não-supervisionado que, em conjunto com técnicas de Aprendizado Ativo e Aprendizado por Transferência, reduz a necessidade de rótulos verdadeiros para manter boa performance de um classificador ao longo do tempo, mesmo com a ocorrência de mudanças de conceito. / Despite the relatively maturity of batch-mode supervised learning research, in which the data typifies stationary problems, many real world applications deal with data streams whose statistical distribution changes over time, causing what is known as concept drift. A large body of research has been done in the last years, with the objective of creating new models that are accurate even in the presence of concept drifts. However, most of them assume that, once the classification algorithm labels an event, its actual label become readily available. This work explores the complementary situations, with a review of the most important published works and an analysis over the impact of delayed true labeling, including no true label availability at all. Furthermore, this work proposes a new algorithm that heavily reduces the complexity of applying Kolmogorov- Smirnov non-parametric hypotheis test, turning it into an uselful tool for analysis on data streams. As an instantiation of its usefulness, we present an unsupervised drift-detection method that, along with Active Learning and Transfer Learning approaches, decreases the number of true labels that are required to keep good classification performance over time, even in the presence of concept drifts.
|
556 |
Untersuchung der Driftinstabilität an der rotierenden magnetisierenden Plasmasäule des PSI-1 im Falle eines Plasmahohlprofils und großer endlicher Ionengyroradieneffekte / Investigation of the drift instability on the rotating magnetized plasma column of the PSI-1 in the case of a hollow plasma profile and large finite ion gyroradii effectsKlose, Sören 22 June 2000 (has links)
Es werden die Eigenschaften von Driftwellen untersucht, welche an der rotierenden magnetisierten Plasmasäule des PSI-1/2 beobachtet wurden. Ihre parallelen Wellenzahlen sind sehr klein, sie besitzen eine nahezu lineare azimutale Dispersionsrelation, ihre azimutale Phasengeschwindigkeit ist etwa gleich der azimutalen Ionendriftgeschwindigkeit und die Potentialfluktuationen eilen den Dichtefluktuationen um etwa Pi hinterher. Deshalb können sie als Ionendriftwellen bezeichnet werden. Alle zugehörigen Messungen wurden mit Langmuirsonden, Spektrometern und schnellen CCD-Kameras durchgeführt. Ein einfaches analytisches Modell, welches die Potential- und Dichteverteilung der Ionendriftwelle beschreibt, offenbart den zugrundeliegenden Instabilitätsmechanismus. Weiterhin wird eine Klassifikation der beobachteten Ionendriftinstabilität in einem Übersichtsschema der bekannten Driftinstabilitäten gegeben. Abschließend wurden die parallele und senkrechte Teilchendiffusion untersucht. Während der gemessene parallele Diffusionskoeffizient sehr gut mit den theoretischen Vorstellungen übereinstimmt, kann die beobachtete Senkrechtdiffusion nicht allein durch die klassische Diffusion erklärt werden. Email: sok@ipp.mpg.de / The properties of drift waves observed in the rotating magnetized plasma column of the PSI-1/2 will be shown. Their parallel wave numbers are very small, their azimuthal dispersion relation is approximately linear, their resultant azimuthal phase velocity is nearly equal to the azimuthal ion drift velocity, and the potential fluctuations always lag behind the density fluctuations with a phase shift of approximately Pi. Therefore, they can be identified as ion drift waves. All corresponding measurements were performed with Langmuir-probes, spectrometers and fast CCD-cameras. A simple analytical model, describing the potential- and density distribution of the ion drift wave, shows the underlying mechanism of the instability. Furthermore, a classification of the observed ion drift instability is given, by referring to a survey scheme summarizing all known drift instabilities. Finally, the parallel and perpendicular particle diffusion were investigated. While the measured parallel diffusion coefficient is in a good agreement with the theory, the observed perpendicular diffusion can not be described with the classical diffusion alone. Email: sok@ipp.mpg.de
|
557 |
Aging studies of drift chambers of the HERA-B outer tracker using CF 4 -based gasesSchreiner, Alexander 15 November 2001 (has links)
ÿþD / The intense radiation environment in the HERA-B experiment, being comparable to that of LHC experiments, requires that the detector be very resistant to high radiation loads. The Outer Tracker of HERA-B consists of drift tubes folded from polycarbonate foils (honeycomb) and is operated with a CF4-containing gas mixture. Aging tests made in HERA-B, under hadronic irradiation, using prototype drift chambers showed a rapid rise of self-sustaining dark currents (Malter effect). Extensive aging studies were carried out with the objective to find an appropriate set of construction materials as well as optimal operational parameters. It was shown that the gain loss (anode aging) and the Malter effect could be avoided after replacing the methane in the counting gas, Ar/CF4/CH4, by CO2 and applying cathode coating. However, other, relatively less known aging effects appeared: rising conductivity of the wire-supporting strips and wire corrosion. These two aging effects could also be prevented keeping water at a proper concentration. The aging tests showed that a too high water content (above about 500 ppm) gives rise to strip conductivity, conversely, a too low water content (below about 100 ppm) enhances anode wire corrosion. In an attempt to interpret the test results, plasma phenomena like quenching by different admixtures, production of molecular species, concentration of long-lived HF produced via CF4 dissociation in the avalanche were semi-quantitatively estimated. This work contains also a comprehensive summary of all aging studies for the Outer Tracker and an overview of the literature on wire chamber aging, in particular with CF4-based counting gas mixtures.
|
558 |
Medidas de velocidade de arrastamento de elétrons em gases inibidores de descargas pelo método de Townsend pulsado / Measurements of electron drift velocity in quenching gases using the pulsed Townsend techniqueVivaldini, Túlio Cearamicoli 18 June 2014 (has links)
O deslocamento de elétrons em gases pode ser caracterizado por grandezas macroscópicas tais como a velocidade de arrastamento, taxa de ionização, primeiro coeficiente de Townsend e os coeficientes de difusão longitudinal e transversal, que são denominados de parâmetros de transporte. Esses parâmetros são importantes já que permitem a validação das secções de choque de colisões de elétrons com as moléculas do gás, contribuem com informações para modelos de descargas automantidas e são importantes para o desenvolvimento de novos detectores gasosos. Neste trabalho são apresentados os resultados de velocidade de arrastamento (W), taxa de ionização (Ri) e do primeiro coeficiente Townsend de ionização (α), para o isobutano e n-butano em função do campo elétrico reduzido efetivo no intervalo de 130 a 180 Td, obtidos utilizando a técnica de Townsend pulsada. Em nosso aparato, os elétrons primários são gerados a partir da incidência no catodo de um curto pulso de laser e são acelerados em direção ao anodo por meio de um campo elétrico uniforme. O sinal elétrico proveniente do deslocamento desses elétrons é digitalizado por um osciloscópio e a partir do ajuste de uma função modelo à forma do sinal elétrico, determinam-se os parâmetros de transporte. Os resultados obtidos para o isobutano e n-butano foram comparados com os valores da simulação Magboltz 2-versão 8.6 em razão da escassez de dados na literatura para esses gases na região de campo elétrico reduzido estudada. / The electron swarm can be characterized by macroscopic quantities such as the drift velocity, ionization rate, first Townsend ionization coefficient and the longitudinal and transverse diffusion coefficients, so called transport parameters. These parameters are important since they allow the validation of electron collisional cross section with gas molecules, contribute for self-sustained discharge models and are important for gaseous detectors development. In the present work the results of electron drift velocity (W), ionization rate (Ri) and first Townsend coefficient (α) as a function of the reduced electric field for isobutane and n-butane by the means of pulsed Townsend technique are presented. In our experimental setup, the primary electrons are liberated by the irradiation of a short laser pulse at the cathode and are accelerated towards the anode through a uniform electric field. The electron movements induce signals that are digitalized and the fitting of a model function to their waveforms provides the transport parameters. The results obtained for isobutane and n-butane were compared with Magboltz 2-version 8.6 values, since there are few data in the literature for these gases for effective reduced electric field ranging from 130 to 180 Td.
|
559 |
Evolução morfológica no crânio de Lemuriformes (Primates:Strepsirrhini) / Lemuriformes (Primates: Strepsirrhini) cranial morphological evolutionPenna, Anna Paula Casselli 27 July 2016 (has links)
O principal objetivo deste estudo foi investigar os processos responsáveis pela evolução morfológica craniana de primatas da subordem Strepsirrhini, com maior ênfase ao clado endêmico à ilha de Madagascar. Sistemas multidimensionais morfológicos como o crânio descrevem a quantidade de variação disponível para atuação de processos evolutivos. A partir de 27 marcos anatômicos cranianos e 39 distâncias euclidianas estimei matrizes de variância e covariância fenotípicas (Matrizes P) para 40 espécies das sete famílias viventes de Strepsirrhini e duas espécies do gênero Tarsius. Utilizei a abordagem da genética quantitativa para investigar o padrão de variação nas matrizes P dessa linhagem de primatas, conjuntamente com um par de matrizes fenotípica e genética de Saguinus fuscicolis. Para investigar a estabilidade e conservção do padrão da estrutura de variação ao longo da diversidade deste clado de primatas compararei estas matrizes através de dois métodos complementares que me permitiram avaliar a distribuição da variação no morfoespaço (Random Skewers) e o compartilhamento de direções de variação (Krzanowski) das matrizes. Os resultados obtidos sugerem uma estabilidade das matrizes, possivelmente mantida por seleção estabilizadora interna comum ao clado, aos demais primatas e mamíferos em geral. Esses resultados de comparação de matrizes são apresentados no Capítulo 1 e corroboram a conjectura de Cheverud, indicando que as matrizes cumprem o pressuposto de similaridade estrutural entre as matrizes P e a matriz G. Além disso o padrão de variação apresentou um sinal filogenético e algumas das diferenças observadas nas matrizes parecem estar relacionadas com fatores de dieta. No Capítulo 2 investiguei a evolução da divergência morfológica craniana em um conjunto de 70 espécies, onde testei a hipótese nula de deriva genética como explicação suficiente para a divergencia entre as médias das espécies. Esta hipótese foi avaliada dentro das expectativas de modelos da genética quantitativa através de duas metodologias. Tais testes avaliam a proporcionalidade de variação dentro (W) e entre grupos (B) sob perspectivas complementares. Com o teste de regressão investiguei a manutenção da quantidade de variação associada a cada eixo de variação, e com o teste de correlação avaliei o alinhamento entre estes eixos. Rejeitamos a H_0 em diversos pontos da árvore que representam eventos de cladogenese a nível de família e gênero. Nestes casos aceitamos a H_1 de seleção natural e vemos que a maior parte da diversificação da linhagem de primatas de Madagascar foi influenciada por seleção direcional envolvendo tamanho e alterações anatômicas relacionadas a dieta / The main goal of this study was to investigate the evolutionary processes responsible for Strepsirrhini cranial morphological evolution. Multidimensional morphological structures like the cranium can describe the amount of variance available to evolution. We used 27 landmarks and 39 euclidean distances between them to describe the variation in the cranium. We used a broad phylogenetic scaled comparison of cranial P matrices representing 40 species of Strepsirrhini primates combined to a pair of Saguinus fuscicolis G and P to investigate the stability of variance structure along this lineage. Our results show a relative stability in the patterns of variance and covariance of the clade and that P matrices can be considered as surrogates to its underling G. We also report a high association between matrix similarity and phylogenetic distance. We investigated particular contributions of each trait to matrix dissimilarity through a evolutionary perspective considering the main dietary shifts observed in the clade. Our results suggest that differences in observed patterns of variation can be attributed to characters with heterogeneity in the degree of stabilizing selection in the adult cranium and to differences in directional selection involved in chewing of specialized feeding behavior. Under the quantitative genetics theory lies an expectation for the evolution of average phenotypes that if populations have diverged by random processes, patterns of within and between-taxon morphological variation should be proportional. In chapter 2 we tested the null hypothesis that genetic drift is a sufficient explanation to observed divergence in cranial multivariate means of 70 Strepsirrhini primates. We detected deviations from neutrality along the whole clade of Strepsirrhini and specially in more inclusive nodes. We argue that this deviations reflect important historical shifts in the evolution of the clade associated with directional selection for size and in anatomical features related with diet
|
560 |
Classificação de fluxos de dados não estacionários com algoritmos incrementais baseados no modelo de misturas gaussianas / Non-stationary data streams classification with incremental algorithms based on Gaussian mixture modelsOliveira, Luan Soares 18 August 2015 (has links)
Aprender conceitos provenientes de fluxos de dados é uma tarefa significamente diferente do aprendizado tradicional em lote. No aprendizado em lote, existe uma premissa implicita que os conceitos a serem aprendidos são estáticos e não evoluem significamente com o tempo. Por outro lado, em fluxos de dados os conceitos a serem aprendidos podem evoluir ao longo do tempo. Esta evolução é chamada de mudança de conceito, e torna a criação de um conjunto fixo de treinamento inaplicável neste cenário. O aprendizado incremental é uma abordagem promissora para trabalhar com fluxos de dados. Contudo, na presença de mudanças de conceito, conceitos desatualizados podem causar erros na classificação de eventos. Apesar de alguns métodos incrementais baseados no modelo de misturas gaussianas terem sido propostos na literatura, nota-se que tais algoritmos não possuem uma política explicita de descarte de conceitos obsoletos. Nesse trabalho um novo algoritmo incremental para fluxos de dados com mudanças de conceito baseado no modelo de misturas gaussianas é proposto. O método proposto é comparado com vários algoritmos amplamente utilizados na literatura, e os resultados mostram que o algoritmo proposto é competitivo com os demais em vários cenários, superando-os em alguns casos. / Learning concepts from data streams differs significantly from traditional batch learning. In batch learning there is an implicit assumption that the concept to be learned is static and does not evolve significantly over time. On the other hand, in data stream learning the concepts to be learned may evolve over time. This evolution is called concept drift, and makes the creation of a fixed training set be no longer applicable. Incremental learning paradigm is a promising approach for learning in a data stream setting. However, in the presence of concept drifts, out dated concepts can cause misclassifications. Several incremental Gaussian mixture models methods have been proposed in the literature, but these algorithms lack an explicit policy to discard outdated concepts. In this work, a new incremental algorithm for data stream with concept drifts based on Gaussian Mixture Models is proposed. The proposed methodis compared to various algorithms widely used in the literature, and the results show that it is competitive with them invarious scenarios, overcoming them in some cases.
|
Page generated in 0.0326 seconds