Comprendre la genèse d’une situation de conduite requiert d’analyser les choixfaits par le conducteur au volant de son véhicule pendant l’activité de conduite, dans sacomplexité naturelle et dans sa dynamique située. Le LESCOT a développé le modèleCOSMODRIVE, fournissant un cadre conceptuel pour la simulation cognitive de l’activitéde conduite automobile. Pour exploiter ce modèle en simulation, il est nécessairede produire les connaissances liées à la situation de conduite sous forme d’un automatepar exemple. La conception d’un tel automate nécessite d’une part de disposer de donnéesissues de la conduite réelle, enregistrées sur un véhicule instrumenté et d’autrepart d’une expertise humaine pour les interpréter.Pour accompagner ce processus d’ingénierie des connaissances issues de l’analysed’activité, ce travail de thèse propose une méthode de découverte interactive deconnaissances à partir de traces d’activité. Les données de conduite automobile sontconsidérées comme des M-Traces, associant une sémantique explicite aux données,exploitées en tant que connaissances dans un Système à Base de Traces (SBT). Le SBTpermet de filtrer, transformer, reformuler et abstraire les séquences qui serviront à alimenterla synthèse de modèles automates de l’activité de conduite. Nous reprenons destechniques de fouille de workflow permettant de construire des automates (réseaux dePetri) à partir de logs. Ces techniques nécessitent des données complètes ou statistiquementreprésentatives. Or les données collectées à bord d’un véhicule en situationde conduite sont par nature des cas uniques, puisqu’aucune situation ne sera jamaisreproductible à l’identique, certaines situations particulièrement intéressantes pouvanten outre être très rarement observées. La gageure est alors de procéder à une forme degénéralisation sous la forme de modèle, à partir d’un nombre de cas limités, mais jugéspertinents, représentatifs, ou particulièrement révélateurs par des experts du domaine.Pour compléter la modélisation de telles situations, nous proposons donc de rendreinteractifs les algorithmes de synthèse de réseau de Petri à partir de traces, afin depermettre à des experts-analystes de guider ces algorithmes et de favoriser ainsi la découvertede connaissances pertinentes pour leur domaine d’expertise. Nous montreronscomment rendre interactifs l’algorithme α et l’algorithme α+ et comment généralisercette approche à d’autres algorithmes.Nous montrons comment l’utilisation d’un SBT et de la découverte interactived’automates impacte le cycle général de découverte de connaissances. Une méthodologieest proposée pour construire des modèles automates de l’activité de conduiteautomobile.Une étude de cas illustre la méthodologie en partant de données réelles de conduiteet en allant jusqu’à la construction de modèles avec un prototype logiciel développédans le cadre de cette thèse / Driving is a dynamic and complex activity. Understanding the origin of a driving situationrequires the analysis of the driver’s choices made while he/she drives. In addition,a driving situation has to be studied in its natural complexity and evolution. LESCOThas developed a model called COSMODRIVE, which provides a conceptual frameworkfor the cognitive simulation of the activity of car driving. In order to run themodel for a simulation, it is necessary to gather knowledge related to the driving situation,for example in the form of an automaton. The conception of such an automatonrequires : 1) the use of real data recorded in an instrumented car, and, 2) the use of humanexpertise to interpret these data. These data are considered in this thesis as activitytraces.The purpose of this thesis is to assist the Knowledge Engineering process of activityanalysis. The present thesis proposes a method to interactively discover knowledgefrom activity traces. For this purpose, data from car driving are considered as M-traces– which associate an explicit semantic to these data. This semantic is then used asknowledge in a Trace Based System. In a Trace Based System, M-traces can be filtered,transformed, reformulated, and abstracted. The resulting traces are then used as inputsin the production of an automaton model of the activity of driving. In this thesis,Workflow Mining techniques have been used to build automata (Petri nets) from logs.These techniques require complete or statistically representative data sets. However,data collected from instrumented vehicles are intrinsically unique, as no two drivingsituations will ever be identical. In addition, situations of particular interest, such ascritical situations, are rarely observed in instrumented vehicle studies. The challenge isthen to produce a model which is a form of generalisation from a limited set of cases,which have been judged by domain experts as being relevant and representative of whatactually happens.In the current thesis, algorithms synthesising Petri nets from traces have been madeinteractive, in order to achieve the modelling of such driving situations. This thenmakes it possible for experts to guide the algorithms and therefore to support the discoveryof knowledge relevant to the experts. The process involved in making the α-algorithm and the α+-algorithm interactive is discussed in the thesis in a way that canbe generalised to other algorithms.In addition, the current thesis illustrates how the use of a Trace Based System andthe interactive discovery of automata impacts the global cycle of Knowledge Discovery.A methodology is also proposed to build automaton models of the activity of cardriving. Finally, a case study is presented to illustrate how the proposed methodologycan be applied to real driving data in order to construct models with the softwaredeveloped in this thesis

Made available in DSpace on 2014-07-29T15:08:16Z (GMT). No. of bitstreams: 1 Tese de Kedma Batista Duarte.pdf: 2986348 bytes, checksum: b08f936c5937365d2b7493f4db0f0b88 (MD5) Previous issue date: 2010-12-20 / The Genetic Programming (GP) is a heuristic algorithm for Data Mining (DM), which can be applied to the classification task. This is a method of evolutionary computing inspired in the mechanisms of natural selection theory of Charles Darwin, declared in 1859 in his book "The Origin of Species." From an initial population, the method search over a number of generations to find solutions adapted to the environment of problem. The PG method was proposed in 1990 by John Koza, who demonstrated in one of its applications, the induction in formation of decision trees in the process of data classification. Within this context, the study developed in this work has as main objective the investigation of the concepts of PG and its application on a database of scientific collaboration networks, helping as a management tool in prospective studies of trends for the establishment of common axes in public policy of Science, Technology and Innovation (STI), focusing on regional development. The method is applied on a set of attributes, sorting them in order to identify similarity relationships between groups of researchers that comprise the network. The study involves the concepts of Knowledge Discovery in Databases (KDD) and Data Mining (DM). Networks of Scientific Collaboration, or Networks Research, are inserted in the context of small groups of social networks, the environment is dynamic due to the easy of information exchange and links between individuals, favoring the formation of new groups, which makes the growth of the network unlimited. "The combination of these groups, generated by the relationships between them, appears as a case of multi-criteria decision, granting the application of some complexity. In this sense, it is intended to apply the method of PG for generation of classification rules that lead to the discovery of groups of researchers with similar traits, which in a planned process could be induced to form groups strengthened and consolidated. The study helps to exploit the potential of genetic programming as a classifier algorithm, as well as use it as a method to build tools to support planning and decision making in STI. / A Programação Genética (PG) é um algoritmo heurístico de Mineração de Dados (MD), quepode ser aplicado na tarefa de classificação. Trata-se de um método da Computação Evolutiva inspirado nos mecanismos de seleção natural, da teoria de Charles Darwin, declarada em 1859 em seu livro A Origem das Espécies . A partir de uma população inicial, o método busca ao longo de um conjunto de gerações a descoberta de soluções bem adaptadas ao ambiente do problema. O método de PG foi proposto por John Koza em 1990, que demonstrou em uma de suas aplicações, a indução na formação de árvores de decisão em processos de classificação de dados. Dentro deste contexto, o estudo desenvolvido neste trabalho tem como objetivo principal a investigação dos conceitos de PG e sua aplicação sobre uma base de dados de Redes de Colaboração Científica, auxiliando como ferramenta de gestão em estudos prospectivos de tendências para o estabelecimento de eixos comuns em políticas públicas de Ciência, Tecnologia e Inovação (CT&I), com foco em desenvolvimento regional. O método é aplicado sobre um conjunto de atributos, classificando-os de forma a identificar relações de similaridade entre os grupos de pesquisadores que compõem a rede. O estudo envolve conceitos de Descoberta do Conhecimento em Bases de Dados (DCBD) e Mineração de Dados (MD). As Redes de Colaboração Científica, ou Redes de Pesquisa, estão inseridas no contexto dos pequenos grupos das Redes Sociais, o ambiente é dinâmico devido à facilidade para troca de informações e articulação entre os indivíduos, favorecendo a formação de novos grupos, fato que torna ilimitado o crescimento da Rede. A combinação das características desses grupos, gerada pelos relacionamentos entre eles, configura-se como um caso de decisão multi-critério, dotando a aplicação de certa complexidade. Neste sentido, pretende-se com a aplicação do método da PG a geração de regras de classificação que levem à descoberta de grupos de pesquisadores com características similares, que em um processo planejado poderiam ser induzidos à formação de grupos fortalecidos e consolidados. O estudo contribui no sentido de explorar o potencial da Programação Genética como um algoritmo classificador, bem como, usá-lo como método na construção de ferramentas de apoio ao planejamento e tomada de decisão em CT&I.