京都大学 / 新制・課程博士 / 博士(理学) / 甲第25096号 / 理博第5003号 / 京都大学大学院理学研究科数学・数理解析専攻 / (主査)准教授 照井 一成, 教授 大木谷 耕司, 教授 長谷川 真人 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

A huge amount of data is generated daily by digital technologies such as social media, web logs, traffic sensors, on-line transactions, tracking data, videos, and so on. This has led to the archiving and storage of larger and larger datasets, many of which are multi-modal, or contain different types of data which contribute to the problem that is now known as “Big Data”. In the area of Big Data, volume, variety and velocity problems remain difficult to solve. The work presented in this thesis focuses on the variety aspect of Big Data. For example, data can come in various and mixed formats for the same feature(attribute) or different features and can be identified mainly by one of the following data types: real-valued, crisp and linguistic values. The increasing variety and ambiguity of such data are particularly challenging to process and to build accurate machine learning models. Therefore, data heterogeneity requires new methods of analysis and modelling techniques to enable useful information extraction and the modelling of achievable tasks. In this thesis, new approaches are proposed for handling heterogeneous Big Data. these include two techniques for filtering heterogeneous data objects are proposed. The two techniques called Two-Dimensional Similarity Space(2DSS) for data described by numeric and categorical features, and Three-Dimensional Similarity Space(3DSS) for real-valued, crisp and linguistic data are proposed for filtering such data. Both filtering techniques are used in this research to reduce the noise from the initial dataset and make the dataset more homogeneous. Furthermore, a new similarity measure based on intuitionistic fuzzy set theory is proposed. The proposed measure is used to handle the heterogeneity and ambiguity within crisp and linguistic data. In addition, new combine similarity models are proposed which allow for a comparison between the heterogeneous data objects represented by a combination of crisp and linguistic values. Diverse examples are used to illustrate and discuss the efficiency of the proposed similarity models. The thesis also presents modification of the k-Nearest Neighbour classifier, called k-Nearest Neighbour Weighted Average (k-NNWA), to classify the heterogeneous dataset described by real-valued, crisp and linguistic data. Finally, the thesis also introduces a novel classification model, called FCCM (Filter Combined Classification Model), for heterogeneous data classification. The proposed model combines the advantages of the 3DSS and k-NNWA classifier and outperforms the latter algorithm. All the proposed models and techniques have been applied to weather datasets and evaluated using accuracy, Fscore and ROC area measures. The experiments revealed that the proposed filtering techniques are an efficient approach for removing noise from heterogeneous data and improving the performance of classification models. Moreover, the experiments showed that the proposed similarity measure for intuitionistic fuzzy data is capable of handling the fuzziness of heterogeneous data and the intuitionistic fuzzy set theory offers some promise in solving some Big Data problems by handling the uncertainties, and the heterogeneity of the data.

<p>Intuicionistički rasplinuti skupovi su već proučavani i definisani u kontekstu mrežnovrednosnih struktura, ali svaka od postojećih definicija imala je odgovarajuće nedostatke. U ovom radu razvijena je definicija intuicionističkog poset-vrednosnog rasplinutog skupa, kojom se poset predstavlja kao podskup distributivne mreže. Na ovaj način možemo ispitivati funkcije pripadanja i nepripadanja i njihove odnose bez upotrebe komplementiranja na posetu. Takođe, u ovako postavljenim okvirima, svaki poset (a samim tim i mreža) može biti kodomen intuicionističkog rasplinutog skupa (čime se isključuje uslov ograničenosti poseta). Primenom uvedene definicije razmatrane su IP-vrednosne rasplinute relacije, x-blokovi ovih relacija i familije<br />njihovih nivoa.Razvijene su jake poset vrednosne relacije reciprociteta koje&nbsp; predstavljaju uop&scaron;tenje relacija reciprociteta sa intervala [0,1]. Pokazano je da ovakve relacije imaju svojstva slična poset-vrednosnim relacijama preferencije. Međutim, postoje velika ograničenja za primenu ovakvih relacija jer su zahtevi dosta jaki.<br />Uvedene su IP-vrednosne relacije reciprociteta koje se mogu definisati za veliku klasu poseta.Ovakve relacije pogodne su za opisivanje preferencija. Posmatrana je intuicionistička poset-vrednosna relacija preferencije, koja je refleksivna rasplinuta relacija, nad skupom alternativa. U samom procesu vi&scaron;ekriterijumskog odlučivanja<br />može se pojaviti situacija kada alternative nisu međusobno uporedive u odnosu na relaciju preferencije, kao i nedovoljna određenost samih alternativa. Da bi se prevazi&scaron;li ovakvi problemi uvodi se intuicionistička poset-vrednosna relacija preferencije kao intuicionistička rasplinuta relacija na skupu alternativa sa vrednostima u uređenom skupu. Analizirana su neka njena svojstva. Ovakav model pogodan je za upoređivanje alternativa koje nisu, nužno, u linearnom poretku. Dato je nekoliko opravdanja za uvodjenje oba tipa definisanih relacija. Jedna od mogućnosti jeste preko mreže intervala elemenata iz konačnog lanca S, a koji predstavljaju ocene određene alternative. Relacije preferencije mogu uzimati vrednosti sa ove mreže i time se može prevazići nedostatak informacija ili neodlučnost donosioca odluke.</p> / <p>Intuitionistic fuzzy sets have already been explored in depth and defined in the context of lattice-valued intuitionistic fuzzy sets, however, every existing definition has certain drawbacks. In this thesis, a definition of poset-valued intuitionistic fuzzy sets is developed, which introduces a poset as a subset of a distributive lattice. In this manner, functions of membership and non-membership can be examined as well as&nbsp; their relations without using complement in the poset. Also, in such framework, each poset (and the lattice) can be a co-domain of an intuitionistic fuzzy set (which excludes the condition of the bounded poset). Introduced definition defines IP-valued fuzzy relations, x-blocks of these relations andfamilies of their levels. Strong IP-valued&nbsp; reciprocialy relations have been developed as a generalization of reciprocal relations from interval [0,1]. It has been shown that these relations have properties similar to the P-valued preferences relations. However, there are great constraints on the application of these relations because the requirements are quite strong.IP- valued reciprocial relations have been introduced, which can be defined for a large class of posets. Such relations are suitable for describing preferences.An intuitionistic poset-valued preference relation, which is a reflexive fuzzy relation, over the set of&nbsp; alternatives, has been examined. In the process of a multi-criteria decision making, a situation can occur that the alternatives cannot be compared by the preference relation, as well as insufficient determination of the mentioned alternatives. In order to overcome similar problems, we have introduced an intuitionistic poset-valued preference relation as an intuitionistic fuzzy set over the set of alternatives with values in a certain poset. We have analyzed some its performances. This model is suitable for comparing alternatives which are not necessarily linearly ordered. There are several justifications for the introduction of&nbsp; both types of defined relations. One of the possibilities is via the lattice of the intervals&nbsp; of elements from the finite chain S, which represent the preference of a particular alternative. Preferences relations can take values from this lattice and this can overcome the lack of informations or the decisiveness of the decision maker.</p>

Made available in DSpace on 2014-12-17T14:55:08Z (GMT). No. of bitstreams: 1 ClaudileneGC_TESE.pdf: 853804 bytes, checksum: 011bfb4befb8b54fce2cd4b1b724efdb (MD5) Previous issue date: 2012-08-20 / The idea of considering imprecision in probabilities is old, beginning with the Booles George work, who in 1854 wanted to reconcile the classical logic, which allows the modeling of complete ignorance, with probabilities. In 1921, John Maynard Keynes in his book made explicit use of intervals to represent the imprecision in probabilities. But only from the work ofWalley in 1991 that were established principles that should be respected by a probability theory that deals with inaccuracies. With the emergence of the theory of fuzzy sets by Lotfi Zadeh in 1965, there is another way of dealing with uncertainty and imprecision of concepts. Quickly, they began to propose several ways to consider the ideas of Zadeh in probabilities, to deal with inaccuracies, either in the events associated with the probabilities or in the values of probabilities. In particular, James Buckley, from 2003 begins to develop a probability theory in which the fuzzy values of the probabilities are fuzzy numbers. This fuzzy probability, follows analogous principles to Walley imprecise probabilities. On the other hand, the uses of real numbers between 0 and 1 as truth degrees, as originally proposed by Zadeh, has the drawback to use very precise values for dealing with uncertainties (as one can distinguish a fairly element satisfies a property with a 0.423 level of something that meets with grade 0.424?). This motivated the development of several extensions of fuzzy set theory which includes some kind of inaccuracy. This work consider the Krassimir Atanassov extension proposed in 1983, which add an extra degree of uncertainty to model the moment of hesitation to assign the membership degree, and therefore a value indicate the degree to which the object belongs to the set while the other, the degree to which it not belongs to the set. In the Zadeh fuzzy set theory, this non membership degree is, by default, the complement of the membership degree. Thus, in this approach the non-membership degree is somehow independent of the membership degree, and this difference between the non-membership degree and the complement of the membership degree reveals the hesitation at the moment to assign a membership degree. This new extension today is called of Atanassov s intuitionistic fuzzy sets theory. It is worth noting that the term intuitionistic here has no relation to the term intuitionistic as known in the context of intuitionistic logic. In this work, will be developed two proposals for interval probability: the restricted interval probability and the unrestricted interval probability, are also introduced two notions of fuzzy probability: the constrained fuzzy probability and the unconstrained fuzzy probability and will eventually be introduced two notions of intuitionistic fuzzy probability: the restricted intuitionistic fuzzy probability and the unrestricted intuitionistic fuzzy probability / A id?ia de considerar imprecis?o em probabilidades ? antiga, remontando aos trabalhos de George Booles, que em 1854 pretendia conciliar a l?gica cl?ssica, que permite modelar ignor?ncia completa, com probabilidades. Em 1921, John Maynard Keynes em seu livro fez uso expl?cito de intervalos para representar a imprecis?o nas probabilidades. Por?m, apenas a partir dos trabalhos de Walley em 1991 que foram estabelecidos princ?pios que deveriam ser respeitados por uma teoria de probabilidades que lide com imprecis?es. Com o surgimento da teoria dos conjuntos fuzzy em 1965 por Lotfi Zadeh, surge uma outra forma de lidar com incertezas e imprecis?es de conceitos. Rapidamente, come?aram a se propor diversas formas de considerar as id?ias de Zadeh em probabilidades, para lidar com imprecis?es, seja nos eventos associados ?s probabilidades como aos valores das probabilidades. Em particular, James Buckley, a partir de 2003 come?a a desenvolver uma teoria de probabilidade fuzzy em que os valores das probabilidades sejam n?meros fuzzy. Esta probabilidade fuzzy segue princ?pios an?logos ao das probabilidades imprecisas de Walley. Por outro lado, usar como graus de verdade n?meros reais entre 0 e 1, como proposto originalmente por Zadeh, tem o inconveniente de usar valores muito precisos para lidar com incertezas (como algu?m pode diferenciar de forma justa que um elemento satisfaz uma propriedade com um grau 0.423 de algo que satisfaz com grau 0.424?). Isto motivou o surgimento de diversas extens?es da teoria dos conjuntos fuzzy pelo fato de incorporar algum tipo de imprecis?o. Neste trabalho ? considerada a extens?o proposta por Krassimir Atanassov em 1983, que adicionou um grau extra de incerteza para modelar a hesita??o ao momento de se atribuir o grau de pertin?ncia, e portanto, um valor indicaria o grau com o qual o objeto pertence ao conjunto, enquanto o outro, o grau com o qual n?o pertence. Na teoria dos conjuntos fuzzy de Zadeh, esse grau de n?o-pertin?ncia por defeito ? o complemento do grau de pertin?ncia. Assim, nessa abordagem o grau de n?o-pertin?ncia ? de alguma forma independente do grau de pertin?ncia, e nessa diferencia entre essa n?o-pertin?ncia e o complemento do grau de pertin?ncia revela a hesita??o presente ao momento de se atribuir o grau de pertin?ncia. Esta nova extens?o hoje em dia ? chamada de teoria dos conjuntos fuzzy intuicionistas de Atanassov. Vale salientar, que o termo intuicionista aqui n?o tem rela??o com o termo intuicionista como conhecido no contexto de l?gica intuicionista. Neste trabalho ser? desenvolvida duas propostas de probabilidade intervalar: a probabilidade intervalar restrita e a probabilidade intervalar irrestrita; tamb?m ser?o introduzidas duas no??es de probabilidade fuzzy: a probabilidade fuzzy restrita e a probabilidade fuzzy irrestrita e por fim ser?o introduzidas duas no??es de probabilidade fuzzy intuicionista: a probabilidade fuzzy intuicionista restrita e a probabilidade fuzzy intuicionista irrestrita

No description available.

No description available.

In this thesis we develop efficient methods for automated proof search within an important class of mathematical logics. The logics considered are the varying, cumulative and constant domain versions of the first-order modal logics K, K4, D, D4, T, S4 and S5, and first-order intuitionistic logic. The use of these non-classical logics is commonplace within Computing Science and Artificial Intelligence in applications in which efficient machine assisted proof search is essential. Traditional techniques for the design of efficient proof methods for classical logic prove to be of limited use in this context due to their dependence on properties of classical logic not shared by most of the logics under consideration. One major contribution of this thesis is to reformulate and abstract some of these classical techniques to facilitate their application to a wider class of mathematical logics. We begin with Bibel's Connection Calculus: a matrix proof method for classical logic comparable in efficiency with most machine orientated proof methods for that logic. We reformulate this method to support its decomposition into a collection of individual techniques for improving the efficiency of proof search within a standard cut-free sequent calculus for classical logic. Each technique is presented as a means of alleviating a particular form of redundancy manifest within sequent-based proof search. One important result that arises from this anaylsis is an appreciation of the role of unification as a tool for removing certain proof-theoretic complexities of specific sequent rules; in the case of classical logic: the interaction of the quantifier rules. All of the non-classical logics under consideration admit complete sequent calculi. We anaylse the search spaces induced by these sequent proof systems and apply the techniques identified previously to remove specific redundancies found therein. Significantly, our proof-theoretic analysis of the role of unification renders it useful even within the propositional fragments of modal and intuitionistic logic.

No description available.

Doctorat en philosophie et lettres / info:eu-repo/semantics/nonPublished

[pt] Garantir que programas são implementados de forma a cumprir uma especificação é uma questão fundamental em computação, por isso, têm sido propostos vários métodos que almejam provar a correção dos programas. Este trabalho apresenta um método, baseado no isomorfismo de Curry-Howard, que extrai conteúdos computacionais de provas intuicionistas, conhecido como síntese construtiva ou proofs-as-programs. É proposto um processo de síntese construtiva de programas, onde a extração do conteúdo computacional gera um programa em linguagem imperativa a partir de uma prova em lógica intuicionista poli-sortida, cujos axiomas definem os tipos abstratos de dados, sendo utilizado como sistema dedutivo a Dedução Natural. Também é apresentada uma prova de correção, bem como uma prova de completude do método atráves do uso de um sistema com regra ômega (computacional) para a aritmética de Heyting, concluindo com uma demonstração da relação entre o uso da indução finita no lugar da regra ômega computacional no processo de síntese. / [en] One of the main problems in computer science is to assure that programs are implemented in such a way that they satisfy a given specification. There are many studies about methods to prove correctness of programs. This work presents a method, belonging to the constructive synthesis or proofs-as-programs paradigm, that comes from the Curry- Howard isomorphism and extracts the computational contents of intuitionist proofs. The synthesis process proposed produces a program in an imperative language from a proof in many-sorted intuitionist logic, where the axioms define the abstract data types using Natural Deduction as deductive system. It is proved the correctness, as well as the completeness of the method regarding the Heyting arithmetic with ômega-rule(in its computational version). A discussion about the use of the finitary induction instead of computational ômega-rule concludes the work.