Sur les tests lisses d'ajustement dans le context des series chronologiques

La plupart des modèles en statistique classique repose sur une hypothèse sur la distribution des données ou sur une distribution sous-jacente aux données. La validité de cette hypothèse permet de faire de l’inférence, de construire des intervalles de confiance ou encore de tester la fiabilité du modèle. La problématique des tests d’ajustement vise à s’assurer de la conformité ou de la cohérence de l’hypothèse avec les données disponibles. Dans la présente thèse, nous proposons des tests d’ajustement à la loi normale dans le cadre des séries chronologiques univariées et vectorielles. Nous nous sommes limités à une classe de séries chronologiques linéaires, à savoir les modèles autorégressifs à moyenne mobile (ARMA ou VARMA dans le cas vectoriel). Dans un premier temps, au cas univarié, nous proposons une généralisation du travail de Ducharme et Lafaye de Micheaux (2004) dans le cas où la moyenne est inconnue et estimée. Nous avons estimé les paramètres par une méthode rarement utilisée dans la littérature et pourtant asymptotiquement efficace. En effet, nous avons rigoureusement montré que l’estimateur proposé par Brockwell et Davis (1991, section 10.8) converge presque sûrement vers la vraie valeur inconnue du paramètre. De plus, nous fournissons une preuve rigoureuse de l’inversibilité de la matrice des variances et des covariances de la statistique de test à partir de certaines propriétés d’algèbre linéaire. Le résultat s’applique aussi au cas où la moyenne est supposée connue et égale à zéro. Enfin, nous proposons une méthode de sélection de la dimension de la famille d’alternatives de type AIC, et nous étudions les propriétés asymptotiques de cette méthode. L’outil proposé ici est basé sur une famille spécifique de polynômes orthogonaux, à savoir les polynômes de Legendre. Dans un second temps, dans le cas vectoriel, nous proposons un test d’ajustement pour les modèles autorégressifs à moyenne mobile avec une paramétrisation structurée. La paramétrisation structurée permet de réduire le nombre élevé de paramètres dans ces modèles ou encore de tenir compte de certaines contraintes particulières. Ce projet inclut le cas standard d’absence de paramétrisation. Le test que nous proposons s’applique à une famille quelconque de fonctions orthogonales. Nous illustrons cela dans le cas particulier des polynômes de Legendre et d’Hermite. Dans le cas particulier des polynômes d’Hermite, nous montrons que le test obtenu est invariant aux transformations affines et qu’il est en fait une généralisation de nombreux tests existants dans la littérature. Ce projet peut être vu comme une généralisation du premier dans trois directions, notamment le passage de l’univarié au multivarié ; le choix d’une famille quelconque de fonctions orthogonales ; et enfin la possibilité de spécifier des relations ou des contraintes dans la formulation VARMA. Nous avons procédé dans chacun des projets à une étude de simulation afin d’évaluer le niveau et la puissance des tests proposés ainsi que de les comparer aux tests existants. De plus des applications aux données réelles sont fournies. Nous avons appliqué les tests à la prévision de la température moyenne annuelle du globe terrestre (univarié), ainsi qu’aux données relatives au marché du travail canadien (bivarié). Ces travaux ont été exposés à plusieurs congrès (voir par exemple Tagne, Duchesne et Lafaye de Micheaux (2013a, 2013b, 2014) pour plus de détails). Un article basé sur le premier projet est également soumis dans une revue avec comité de lecture (Voir Duchesne, Lafaye de Micheaux et Tagne (2016)).

Design of an expandable manufacturing simulator through the application of object-oriented principles

In analysing manufacturing systems, for either design or operational reasons, failure to account for the potentially significant dynamics could produce invalid results. There are many analysis techniques that can be used, however, simulation is unique in its ability to assess detailed, dynamic behaviour. The use of simulation to analyse manufacturing systems would therefore seem appropriate if not essential. Many simulation software products are available but their ease of use and scope of application vary greatly. This is illustrated at one extreme by simulators which offer rapid but limited application whilst at the other simulation languages which are extremely flexible but tedious to code. Given that a typical manufacturing engineer does not posses in depth programming and simulation skills then the use of simulators over simulation languages would seem a more appropriate choice. Whilst simulators offer ease of use their limited functionality may preclude their use in many applications. The construction of current simulators makes it difficult to amend or extend the functionality of the system to meet new challenges. Some simulators could even become obsolete as users, demand modelling functionality that reflects the latest manufacturing system design and operation concepts. This thesis examines the deficiencies in current simulation tools and considers whether they can be overcome by the application of object-oriented principles. Object-oriented techniques have gained in popularity in recent years and are seen as having the potential to overcome any of the problems traditionally associated with software construction. There are a number of key concepts that are exploited in the work described in this thesis: the use of object-oriented techniques to act as a framework for abstracting engineering concepts into a simulation tool and the ability to reuse and extend object-oriented software. It is argued that current object-oriented simulation tools are deficient and that in designing such tools, object -oriented techniques should be used not just for the creation of individual simulation objects but for the creation of the complete software. This results in the ability to construct an easy to use simulator that is not limited by its initial functionality. The thesis presents the design of an object-oriented data driven simulator which can be freely extended. Discussion and work is focused on discrete parts manufacture. The system developed retains the ease of use typical of data driven simulators. Whilst removing any limitation on its potential range of applications. Reference is given to additions made to the simulator by other developers not involved in the original software development. Particular emphasis is put on the requirements of the manufacturing engineer and the need for Ihe engineer to carrv out dynamic evaluations.

Consuming linked data in supply chains:enabling data visibility via linked pedigrees

The performance of a supply chain depends critically on the coordinating actions and decisions undertaken by the trading partners. The sharing of product and process information plays a central role in the coordination and is a key driver for the success of the supply chain. In this paper we propose the concept of "Linked pedigrees" - linked datasets, that enable the sharing of traceability information of products as they move along the supply chain. We present a distributed and decentralised, linked data driven architecture that consumes real time supply chain linked data to generate linked pedigrees. We then present a communication protocol to enable the exchange of linked pedigrees among trading partners. We exemplify the utility of linked pedigrees by illustrating examples from the perishable goods logistics supply chain.

Classificazione e selezione di tecniche di visualizzazione per Big Data Analytics

La tesi presenta uno studio della libreria grafica per web D3, sviluppata in javascript, e ne presenta una catalogazione dei grafici implementati e reperibili sul web. Lo scopo è quello di valutare la libreria e studiarne i pregi e difetti per capire se sia opportuno utilizzarla nell'ambito di un progetto Europeo. Per fare questo vengono studiati i metodi di classificazione dei grafici presenti in letteratura e viene esposto e descritto lo stato dell'arte del data visualization. Viene poi descritto il metodo di classificazione proposto dal team di progettazione e catalogata la galleria di grafici presente sul sito della libreria D3. Infine viene presentato e studiato in maniera formale un algoritmo per selezionare un grafico in base alle esigenze dell'utente.

“It’s the People That Drive It”: Motivating and Supporting Collaborative Data Use Practices in Three Teacher Education Programs

Thesis (Ph.D.)--University of Washington, 2016-08

INVISIBLE LABOR FOR DATA: INSTITUTIONS, INFRASTRUCTURE, AND VIRTUAL SPACE

Americans are accustomed to a wide range of data collection in their lives: census, polls, surveys, user registrations, and disclosure forms. When logging onto the Internet, users’ actions are being tracked everywhere: clicking, typing, tapping, swiping, searching, and placing orders. All of this data is stored to create data-driven profiles of each user. Social network sites, furthermore, set the voluntarily sharing of personal data as the default mode of engagement. But people’s time and energy devoted to creating this massive amount of data, on paper and online, are taken for granted. Few people would consider their time and energy spent on data production as labor. Even if some people do acknowledge their labor for data, they believe it is accessory to the activities at hand. In the face of pervasive data collection and the rising time spent on screens, why do people keep ignoring their labor for data? How has labor for data been become invisible, as something that is disregarded by many users? What does invisible labor for data imply for everyday cultural practices in the United States? Invisible Labor for Data addresses these questions. I argue that three intertwined forces contribute to framing data production as being void of labor: data production institutions throughout history, the Internet’s technological infrastructure (especially with the implementation of algorithms), and the multiplication of virtual spaces. There is a common tendency in the framework of human interactions with computers to deprive data and bodies of their materiality. My Introduction and Chapter 1 offer theoretical interventions by reinstating embodied materiality and redefining labor for data as an ongoing process. The middle Chapters present case studies explaining how labor for data is pushed to the margin of the narratives about data production. I focus on a nationwide debate in the 1960s on whether the U.S. should build a databank, contemporary Big Data practices in the data broker and the Internet industries, and the group of people who are hired to produce data for other people’s avatars in the virtual games. I conclude with a discussion on how the new development of crowdsourcing projects may usher in the new chapter in exploiting invisible and discounted labor for data.

The MUSE Machine -- an Architecture for Structured Data Flow Computation

Computers employing some degree of data flow organisation are now well established as providing a possible vehicle for concurrent computation. Although data-driven computation frees the architecture from the constraints of the single program counter, processor and global memory, inherent in the classic von Neumann computer, there can still be problems with the unconstrained generation of fresh result tokens if a pure data flow approach is adopted. The advantages of allowing serial processing for those parts of a program which are inherently serial, and of permitting a demand-driven, as well as data-driven, mode of operation are identified and described. The MUSE machine described here is a structured architecture supporting both serial and parallel processing which allows the abstract structure of a program to be mapped onto the machine in a logical way.

Collecting valuable consumer data: Motivating consumers to disclose personal information

Collecting and analyzing consumer data is essential in today’s data-driven business environment. However, consumers are becoming more aware of the value of the information they can provide to companies, thereby being more reluctant to share it for free. Therefore, companies need to find ways to motivate consumers to disclose personal information. The main research question of the study was formed as “How can companies motivate consumers to disclose personal information?” and it was further divided into two subquestions: 1) What types of benefits motivate consumers to disclose personal information? 2) How does the disclosure context affect the consumers’ information disclosure behavior? The conceptual framework consisted of a classification of extrinsic and intrinsic benefits, and moderating factors, which were recognized on the basis of prior research in the field. The study was conducted by using qualitative research methods. The primary data was collected by interviewing ten representatives from eight companies. The data was analyzed and reported according to predetermined themes. The findings of the study confirm that consumers can be motivated to disclose personal information by offering different types of extrinsic (monetary saving, time saving, self-enhancement, and social adjustment) and intrinsic (novelty, pleasure, and altruism) benefits. However, not all the benefits are equally useful ways to convince the customer to disclose information. Moreover, different factors in the disclosure context can either alleviate or increase the effectiveness of the benefits and the consumers’ motivation to disclose personal information. Such factors include the consumer’s privacy concerns, perceived trust towards the company, the relevancy of the requested information, personalization, website elements (especially security, usability, and aesthetics of a website), and the consumer’s shopping motivation. This study has several contributions. It is essential that companies recognize the most attractive benefits regarding their business and their customers, and that they understand how the disclosure context affects the consumer’s information disclosure behavior. The likelihood of information disclosure can be increased, for example, by offering benefits that meet the consumers’ needs and preferences, improving the relevancy of the asked information, stating the reasons for data collection, creating and maintaining a trustworthy image of the company, and enhancing the quality of the company’s website.

Using interaction data for improving the offline and online evaluation of search engines

This thesis investigates how web search evaluation can be improved using historical interaction data. Modern search engines combine offline and online evaluation approaches in a sequence of steps that a tested change needs to pass through to be accepted as an improvement and subsequently deployed. We refer to such a sequence of steps as an evaluation pipeline. In this thesis, we consider the evaluation pipeline to contain three sequential steps: an offline evaluation step, an online evaluation scheduling step, and an online evaluation step. In this thesis we show that historical user interaction data can aid in improving the accuracy or efficiency of each of the steps of the web search evaluation pipeline. As a result of these improvements, the overall efficiency of the entire evaluation pipeline is increased. Firstly, we investigate how user interaction data can be used to build accurate offline evaluation methods for query auto-completion mechanisms. We propose a family of offline evaluation metrics for query auto-completion that represents the effort the user has to spend in order to submit their query. The parameters of our proposed metrics are trained against a set of user interactions recorded in the search engine’s query logs. From our experimental study, we observe that our proposed metrics are significantly more correlated with an online user satisfaction indicator than the metrics proposed in the existing literature. Hence, fewer changes will pass the offline evaluation step to be rejected after the online evaluation step. As a result, this would allow us to achieve a higher efficiency of the entire evaluation pipeline. Secondly, we state the problem of the optimised scheduling of online experiments. We tackle this problem by considering a greedy scheduler that prioritises the evaluation queue according to the predicted likelihood of success of a particular experiment. This predictor is trained on a set of online experiments, and uses a diverse set of features to represent an online experiment. Our study demonstrates that a higher number of successful experiments per unit of time can be achieved by deploying such a scheduler on the second step of the evaluation pipeline. Consequently, we argue that the efficiency of the evaluation pipeline can be increased. Next, to improve the efficiency of the online evaluation step, we propose the Generalised Team Draft interleaving framework. Generalised Team Draft considers both the interleaving policy (how often a particular combination of results is shown) and click scoring (how important each click is) as parameters in a data-driven optimisation of the interleaving sensitivity. Further, Generalised Team Draft is applicable beyond domains with a list-based representation of results, i.e. in domains with a grid-based representation, such as image search. Our study using datasets of interleaving experiments performed both in document and image search domains demonstrates that Generalised Team Draft achieves the highest sensitivity. A higher sensitivity indicates that the interleaving experiments can be deployed for a shorter period of time or use a smaller sample of users. Importantly, Generalised Team Draft optimises the interleaving parameters w.r.t. historical interaction data recorded in the interleaving experiments. Finally, we propose to apply the sequential testing methods to reduce the mean deployment time for the interleaving experiments. We adapt two sequential tests for the interleaving experimentation. We demonstrate that one can achieve a significant decrease in experiment duration by using such sequential testing methods. The highest efficiency is achieved by the sequential tests that adjust their stopping thresholds using historical interaction data recorded in diagnostic experiments. Our further experimental study demonstrates that cumulative gains in the online experimentation efficiency can be achieved by combining the interleaving sensitivity optimisation approaches, including Generalised Team Draft, and the sequential testing approaches. Overall, the central contributions of this thesis are the proposed approaches to improve the accuracy or efficiency of the steps of the evaluation pipeline: the offline evaluation frameworks for the query auto-completion, an approach for the optimised scheduling of online experiments, a general framework for the efficient online interleaving evaluation, and a sequential testing approach for the online search evaluation. The experiments in this thesis are based on massive real-life datasets obtained from Yandex, a leading commercial search engine. These experiments demonstrate the potential of the proposed approaches to improve the efficiency of the evaluation pipeline.

Capability Matrix for Open Data

Part 14: Interoperability and Integration

A Bayesian network dealing with measurements and residuals for system monitoring

International audience

WordPress as a framework for automated data capture, filtering and structuring processes

The Exhibitium Project , awarded by the BBVA Foundation, is a data-driven project developed by an international consortium of research groups . One of its main objectives is to build a prototype that will serve as a base to produce a platform for the recording and exploitation of data about art-exhibitions available on the Internet . Therefore, our proposal aims to expose the methods, procedures and decision-making processes that have governed the technological implementation of this prototype, especially with regard to the reuse of WordPress (WP) as development framework.

Multivalent reuse of web data about temporary art exhibitions: The Exhibitium Project

Our proposal aims to display the analysis techniques, methodologies as well as the most relevant results expected within the Exhibitium project framework (http://www.exhibitium.com). Awarded by the BBVA Foundation, the Exhibitium project is being developed by an international consortium of several research groups . Its main purpose is to build a comprehensive and structured data repository about temporary art exhibitions, captured from the web, to make them useful and reusable in various domains through open and interoperable data systems.

Mixing, Chemical Reactions, and Combustion in Supersonic Flows

Experiments were conducted at the GALCIT supersonic shear-layer facility to investigate aspects of reacting transverse jets in supersonic crossflow using chemiluminescence and schlieren image-correlation velocimetry. In particular, experiments were designed to examine mixing-delay length dependencies on jet-fluid molar mass, jet diameter, and jet inclination.

The experimental results show that mixing-delay length depends on jet Reynolds number, when appropriately normalized, up to a jet Reynolds number of 500,000. Jet inclination increases the mixing-delay length, but causes less disturbance to the crossflow when compared to normal jet injection. This can be explained, in part, in terms of a control-volume analysis that relates jet inclination to flow conditions downstream of injection.

In the second part of this thesis, a combustion-modeling framework is proposed and developed that is tailored to large-eddy simulations of turbulent combustion in high-speed flows. Scaling arguments place supersonic hydrocarbon combustion in a regime of autoignition-dominated distributed reaction zones (DRZ). The proposed evolution-variable manifold (EVM) framework incorporates an ignition-delay data-driven induction model with a post-ignition manifold that uses a Lagrangian convected 'balloon' reactor model for chemistry tabulation. A large-eddy simulation incorporating the EVM framework captures several important reacting-flow features of a transverse hydrogen jet in heated-air crossflow experiment.

Domain-driven classification based on multiple criteria and multiple constraint-level programming for intelligent credit scoring

Extracting knowledge from the transaction records and the personal data of credit card holders has great profit potential for the banking industry. The challenge is to detect/predict bankrupts and to keep and recruit the profitable customers. However, grouping and targeting credit card customers by traditional data-driven mining often does not directly meet the needs of the banking industry, because data-driven mining automatically generates classification outputs that are imprecise, meaningless, and beyond users' control. In this paper, we provide a novel domain-driven classification method that takes advantage of multiple criteria and multiple constraint-level programming for intelligent credit scoring. The method involves credit scoring to produce a set of customers' scores that allows the classification results actionable and controllable by human interaction during the scoring process. Domain knowledge and experts' experience parameters are built into the criteria and constraint functions of mathematical programming and the human and machine conversation is employed to generate an efficient and precise solution. Experiments based on various data sets validated the effectiveness and efficiency of the proposed methods. © 2006 IEEE.