Engineers who develop fundamental models for chemical processes are often unable to estimate all of the model parameters due to problems with parameter identifiability and estimability. The literature concerning these two concepts is reviewed and techniques for assessing parameter identifiability and estimability in nonlinear dynamic models are summarized. Modellers often face estimability problems when the available data are limited or noisy. In this situation, modellers must decide whether to conduct new experiments, change the model structure, or to estimate only a subset of the parameters and leave others at fixed values. Estimating only a subset of important model parameters is a technique often used by modellers who face estimability problems and it may lead to better model predictions with lower mean squared error (MSE) than the full model with all parameters estimated. Different methods in the literature for parameter subset selection are discussed and compared. An orthogonalization algorithm combined with a recent MSE-based criterion has been used successfully to rank parameters from most to least estimable and to determine the parameter subset that should be estimated to obtain the best predictions. In this work, this strategy is applied to a batch reactor model using additional data and results are compared with computationally-expensive leave-one-out cross-validation. A new simultaneous ranking and selection technique based on this MSE criterion is also described. Unfortunately, results from these parameter selection techniques are sensitive to the initial parameter values and the uncertainty factors used to calculate sensitivity coefficients. A robustness test is proposed and applied to assess the sensitivity of the selected parameter subset to the initial parameter guesses. The selected parameter subsets are compared with those selected using another MSE-based method proposed by Chu et al. (2009). The computational efforts of these methods are compared and recommendations are provided to modellers. / Thesis (Master, Chemical Engineering) -- Queen's University, 2011-09-27 10:52:31.588

In recent years, concerns of privacy have become more prominent for social networks. Anonymizing a graph meaningfully is a challenging problem, as the original graph properties must be preserved as well as possible. We introduce a generalization of the degree anonymization problem posed by Liu and Terzi. In this problem, our goal is to anonymize a given subset of vertices in a graph while adding the fewest possible number of edges. We examine different approaches to solving the problem, one of which finds a degree-constrained subgraph to determine which edges to add within the given subset and another that uses a greedy approach that is not optimal, but is more efficient in space and time. The main contribution of this thesis is an efficient algorithm for this problem by exploring its connection with the degree-constrained subgraph problem. Our experimental results show that our algorithms perform very well on many instances of social network data. / Graduate

No Aprendizado de Máquina Supervisionado---AM---o algoritmo de indução trabalha com um conjunto de exemplos de treinamento, no qual cada exemplo é constituído de um vetor com os valores dos atributos e as classes, e tem como tarefa induzir um classificador capaz de predizer a qual classe pertence um novo exemplo. Em geral, os algoritmos de indução baseiam-se nos exemplos de treinamento para a construção do classificador, sendo que uma representação inadequada desses exemplos, bem como inconsistências nos mesmos podem tornar a tarefa de aprendizado difícil. Um dos problemas centrais de AM é a Seleção de um Subconjunto de Atributos---SSA---cujo objetivo é diminuir o número de atributos utilizados na representação dos exemplos. São três as principais razões para a realização de SSA. A primeira razão é que a maioria dos algoritmos de AM, computacionalmente viáveis, não trabalham bem na presença de vários atributos. A segunda razão é que, com um número menor de atributos, o conceito induzido através do classificador pode ser melhor compreendido. E, a terceira razão é o alto custo para coletar e processar grande quantidade de informações. Basicamente, são três as abordagens para a SSA: embedded, filtro e wrapper. A Teoria de Rough Sets---RS---é uma abordagem matemática criada no início da década de 80, cuja principal funcionalidade são os redutos, e será tratada neste trabalho. Segundo essa abordagem, os redutos são subconjuntos mínimos de atributos que possuem a propriedade de preservar o poder de descrição do conceito relacionado ao conjunto de todos os atributos. Neste trabalho o enfoque esta na abordagem filtro para a realização da SSA utilizando como filtro os redutos calculados através de RS. São descritos vários experimentos sobre nove conjuntos de dados naturais utilizando redutos, bem como outros filtros para SSA. Feito isso, os atributos selecionados foram submetidos a dois algoritmos simbólicos de AM. Para cada conjunto de dados e indutor, foram realizadas várias medidas, tais como número de atributos selecionados, precisão e números de regras induzidas. Também, é descrito um estudo de caso sobre um conjunto de dados do mundo real proveniente da área médica. O objetivo desse estudo pode ser dividido em dois focos: comparar a precisão dos algoritmos de indução e avaliar o conhecimento extraído com a ajuda do especialista. Embora o conhecimento extraído não apresente surpresa, pôde-se confirmar algumas hipóteses feitas anteriormente pelo especialista utilizando outros métodos. Isso mostra que o Aprendizado de Máquina também pode ser visto como uma contribuição para outros campos científicos. / In Supervised Machine Learning---ML---an induction algorithm is typically presented with a set of training examples, where each example is described by a vector of feature values and a class label. The task of the induction algorithm is to induce a classifier that will be useful in classifying new cases. In general, the inductive-learning algorithms rely on existing provided data to build their classifiers. Inadequate representation of the examples through the description language as well as inconsistencies in the training examples can make the learning task hard. One of the main problems in ML is the Feature Subset Selection---FSS---problem, i.e. the learning algorithm is faced with the problem of selecting some subset of feature upon which to focus its attention, while ignoring the rest. There are three main reasons that justify doing FSS. The first reason is that most ML algorithms, that are computationally feasible, do not work well in the presence of many features. The second reason is that FSS may improve comprehensibility, when using less features to induce symbolic concepts. And, the third reason for doing FSS is the high cost in some domains for collecting data. Basically, there are three approaches in ML for FSS: embedded, filter and wrapper. The Rough Sets Theory---RS---is a mathematical approach developed in the early 1980\'s whose main functionality are the reducts, and will be treated in this work. According to this approach, the reducts are minimal subsets of features capable to preserve the same concept description related to the entire set of features. In this work we focus on the filter approach for FSS using as filter the reducts obtained through the RS approach. We describe a series of FSS experiments on nine natural datasets using RS reducts as well as other filters. Afterwards we submit the selected features to two symbolic ML algorithms. For each dataset, various measures are taken to compare inducers performance, such as number of selected features, accuracy and number of induced rules. We also present a case study on a real world dataset from the medical area. The aim of this case study is twofold: comparing the induction algorithms performance as well as evaluating the extracted knowledge with the aid of the specialist. Although the induced knowledge lacks surprising, it allows us to confirm some hypothesis already made by the specialist using other methods. This shows that Machine Learning can also be viewed as a contribution to other scientific fields.

Context: Software effort estimation is a very important component of the software development life cycle. It underpins activities such as planning, maintenance and bidding. Therefore, it has triggered much research over the past four decades, including many machine learning approaches. One popular approach, that has the benefit of accessible reasoning, is analogy-based estimation. Machine learning including analogy is known to significantly benefit from feature selection/weighting. Unfortunately feature weighting search is an NP hard problem, therefore computationally very demanding, if not intractable. Objective: Therefore, one objective of this research is to develop an effi cient and effective feature weighting algorithm for estimation by analogy. However, a major challenge for the effort estimation research community is that experimental results tend to be contradictory and also lack reliability. This has been paralleled by a recent awareness of how bias can impact research results. This is a contributory reason why software effort estimation is still an open problem. Consequently the second objective is to investigate research methods that might lead to more reliable results and focus on blinding methods to reduce researcher bias. Method: In order to build on the most promising feature weighting algorithms I conduct a systematic literature review. From this I develop a novel and e fficient feature weighting algorithm. This is experimentally evaluated, comparing three feature weighting approaches with a na ive benchmark using 2 industrial data sets. Using these experiments, I explore blind analysis as a technique to reduce bias. Results: The systematic literature review conducted identified 19 relevant primary studies. Results from the meta-analysis of selected studies using a one-sample sign test (p = 0.0003) shows a positive effect - to feature weighting in general compared with ordinary analogy-based estimation (ABE), that is, feature weighting is a worthwhile technique to improve ABE. Nevertheless the results remain imperfect so there is still much scope for improvement. My experience shows that blinding can be a relatively straightforward procedure. I also highlight various statistical analysis decisions which ought not be guided by the hunt for statistical significance and show that results can be inverted merely through a seemingly inconsequential statistical nicety. After analysing results from 483 software projects from two separate industrial data sets, I conclude that the proposed technique improves accuracy over the standard feature subset selection (FSS) and traditional case-based reasoning (CBR) when using pseudo time-series validation. Interestingly, there is no strong evidence for superior performance of the new technique when traditional validation techniques (jackknifing) are used but is more effi cient. Conclusion: There are two main findings: (i) Feature weighting techniques are promising for software effort estimation but they need to be tailored for target case for their potential to be adequately exploited. Despite the research findings showing that assuming weights differ in different parts of the instance space ('local' regions) may improve effort estimation results - majority of studies in software effort estimation (SEE) do not take this into consideration. This represents an improvement on other methods that do not take this into consideration. (ii) Whilst there are minor challenges and some limits to the degree of blinding possible, blind analysis is a very practical and an easy-to-implement method that supports more objective analysis of experimental results. Therefore I argue that blind analysis should be the norm for analysing software engineering experiments.

La consommation excessive d’alcool concerne environ 20% de la population adulte en France. Il est établi qu’une consommation excessive aigue d’alcool engendre une augmentation de la morbi-mortalité en cas d’infection ou de traumatisme. La consommation chronique d’alcool augmente le risque de cancers et d’infection.Toutes ces conséquences médicales sont intimement liées à une altération des défenses de l’hôte induite par l’alcool. L’ingestion d’alcool et ses métabolites engendre une modification de la flore digestive appelée dysbiose ainsi qu’une augmentation de la perméabilité digestive. De cet effet local découle une augmentation du passage d’endotoxines dans le système veineux portal et systémique. Ainsi l’ensemble des éléments impliqués dans la réponse immunologique sont impactés de façon locale (Foie) et systémique par cette endotoxinémie chronique. Deux axes de recherche ont été privilégiés dans cette thèse: La translocation bactérienne et les sous-populations monocytaires. L’originalité de ce travail est l’étude de la cinétique des modifications immunologiques, chez des sujets alcoolo-dépendants (AD) hospitalisés pour un sevrage en alcool durant 2 à 6 semaines.La translocation bactérienne (TB) a été étudiée par l’analyse itérative de marqueurs sériques témoins de celle-ci: l’ADN 16S ribosomial en PCR temps réel, les taux sériques de LBP (LPS-Binding-Protein) et de CD14 soluble par ELISA. Les prélèvements veineux ont été effectués à jeun chez des sujets AD à J0 de leur sevrage et après 4 puis 6 semaines de sevrage. Avant le sevrage (J0) les 3 marqueurs sont plus élevés que dans une population témoin (p<0.001). Après 6 semaines de sevrage, les taux de LBP (p=0.04) et sCD14 (p=0.001) diminuent de manière significative sans revenir aux taux de la population témoin. La consommation de cannabis dans le mois précédent le sevrage est associée à une plus grande diminution des marqueurs LBP et sCD14 au cours du sevrage en alcool.Notre étude confirme une majoration de la TB chez les sujets AD non sevrés et l’impact potentiel du cannabis sur celle-ci. Elle montre également que le délai de sevrage de 6 semaines ne permet pas un retour à la normale des marqueurs de la TB.La répartition, le phénotype et la fonctionnalité des sous-populations monocytaires sanguines ont été étudiés à J0 et après 14 jours de sevrage en alcool chez des sujets AD. Avant le sevrage (J0), la fréquence des monocytes classiques (CD14+CD16-) est diminuée alors que celle des non classiques (CD14dimCD16+) est augmentée chez les sujets AD comparativement aux sujets contrôles. La fréquence des monocytes exprimant les TLR-2 et -4 est réduite chez les sujets AD. La sécrétion basale des cytokines IL-1, IL-6 et TNF est comparable chez les sujets AD et contrôles. En revanche après stimulation in vitro des monocytes par les ligands des TLR-2 et 4, respectivement peptidoglycanes (PGN) et lipopolysaccharides (LPS), la sécrétion d’IL-6 et de TNF est augmentée chez les sujets AD. Le sevrage de 14 jours restaure partiellement la distribution des sous-populations monocytaires. Nos résultats indiquent que la consommation chronique d’alcool altère la distribution, le phénotype et la fonctionnalité des monocytes chez les sujets AD, ces altérations s’améliorent en 14 jours de sevrage mais ne reviennent pas à la normale.Mon travail de thèse confirme l’impact de la consommation chronique d’alcool sur la translocation bactérienne et sur la réponse immune chez l’homme. Il précise la nature et la cinétique de cet impact. Mes résultats suggèrent également que le délai de retour à la normale semble être supérieur à 6 semaines. De nouvelles études permettront de mettre en évidence une cinétique plus précise de ces améliorations biologiques. Les résultats permettent d’envisager des recommandations cliniques quant à une durée d’abstinence pré-opératoire en cas de chirurgie programmée par exemple. / Excessive alcohol consumption concerns about 20% of the French adult population. An acute excessive alcohol consumption named “binge drinking” causes an increase in mortality and morbidity in case of trauma or infection. Chronic alcohol consumption causes an increased risk and severity of infections and increased risk of cancer. All these medical consequences are linked to changes in host defense induced by alcohol consumption. Indeed, alcohol and its metabolites generate modifications of the gut microbiota and increased gut permeability. This local effect causes an increase in endotoxin translocation into portal and systemic blood. Thus, all elements involved in the immune response, and in particular the monocytes as cellular component acting as first line of defense of the immune system, are affected by this chronic endotoxemia. Two research axes were studied during my thesis: Microbial translocation and monocyte subsets. The originality of this work was to study the kinetic of immunological changes induced by alcohol withdrawal in alcohol-dependent (AD) subjects hospitalized for an alcohol withdrawal during 2-6 weeks.Microbial translocation (MT) was studied by iterative analysis of serum markers: Bacterial 16S rDNA levels were measured using qPCR, Lipopolysaccharides-Binding protein (LBP) and soluble CD14 were quantified using ELISA. Blood samples were collected in fasten AD subjects at D0 of alcohol withdrawal and after 4 and 6 weeks of alcohol withdrawal. At D0, the 3 markers of MT were higher in AD subjects than in healthy controls (P<0.001). After 6 weeks of alcohol withdrawal, the serum level of LBP (p=0.04) and sCD14 (p=0.001) were significantly decreased but did not reach rates of healthy controls (HC). Cannabis use during the last month before alcohol withdrawal was associated with a greater decrease in sCD14 and LBP upon alcohol withdrawal. My study confirms an increase in MT in AD subjects and the potential impact of cannabis on this phenomenon, and shows that a 6-weeks abstinence period is not sufficient to return to normal blood biological parameters.Whether and how blood monocyte subsets were impaired in AD patients were studied as well as their evolution after alcohol withdrawal. The CD14+CD16- subset was decreased whereas the CD14dimCD16+ subset was expanded (p<0,001) in AD compared to HC. The frequencies of TLR2- and TLR4-expressing monocytes were reduced in AD compared to HC. Although the basal production of IL-1, IL-6 and TNF by monocytes in AD was comparable to HC, the PGN- and LPS-mediated IL-6 and TNF production were increased in AD. Frequencies of IL-6-expressing monocytes were higher in AD than HC. Alcohol withdrawal partially restored the distribution of monocyte subsets and the frequency of IL-6-producing monocytes, and increased the frequency of TNF-producing cells in response to LPS and PGN stimulation to levels comparable to those in HC. Our findings indicate that chronic alcohol use alters the distribution as well as the phenotypic and functional characteristics of blood monocyte subsets, which are partially restored following 2 weeks of alcohol withdrawal. These studies confirm and specify the impact of chronic alcohol consumption on microbial translocation and on the immune response. Our results also suggest that a period superior to 6 weeks is necessary to reach normal biological parameters. News studies will be necessary to specify the exact kinetic of these biological improvements.These results allow to consider clinical recommendations for a period of abstinence before programmed surgery for example.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22830号 / 医博第4669号 / 新制||医||1047(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 江藤 浩之, 教授 武藤 学, 教授 伊藤 貴浩 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第19106号 / 情博第552号 / 新制||情||98(附属図書館) / 32057 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 河原 達也, 教授 田中 利幸, 講師 吉井 和佳 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第19365号 / 医博第4042号 / 新制||医||1011(附属図書館) / 32379 / 新制||医||1011 / 京都大学大学院医学研究科医学専攻 / (主査)教授 椛島 健治, 教授 佐藤 俊哉, 教授 山田 泰広 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

The need for broadcast encryption arises when a sender wishes to securely distribute messages to varying subsets of receivers, using a broadcast channel, for instance in a pay-TV scenario. This is done by selecting subsets of users and giving all users in the same subset a common decryption key. The subsets will in general be overlapping so that each user belongs to many subsets and has several different decryption keys. When the sender wants to send a message to some users, the message is encrypted using keys that those users have. In this thesis we describe some broadcast encryption schemes that have been proposed in the literature. We focus on stateless schemes which do not require receivers to update their decryption keys after the initial keys have been received; particularly we concentrate on the Subset Difference (SD) scheme. We consider the effects that the logical placement of the receivers in the tree structure used by the SD scheme has on the number of required transmissions for each message. Bounds for the number of required transmissions are derived based on the adjacency of receivers in the tree structure. The tree structure itself is also studied, also resulting in bounds on the number of required transmissions based on the placement of the users in the tree structure. By allowing a slight discrepancy between the set of receivers that the sender intends to send to and the set of receivers that actually can decrypt the message, we can reduce the cost in number of transmissions per message. We use the concept of distortion to quantify the discrepancy and develop three simple algorithms to illustrate how the cost and distortion are related. / <p>Report code: LIU-Tek-Lic-2005:70.</p>