Ce projet de thèse n’a jamais eu de sujet de recherche à proprement parler, mais plutôt un objet, un point d’ancrage, une attache géographique que j’ai choisi pour graviter autour: Kaliningrad. Anciennement partie de la Prusse Orientale, la région revient à l’URSS en 1945 à l’issue de la Seconde guerre mondiale, lorsque cette province allemande est divisée entre trois pays: l’URSS, la Lituanie et la Pologne. Sa capitale, Königsberg, est alors renommée Kaliningrad, en référence à Mikhaïl Kalinine, un des collaborateurs de Staline. En 1991, quand le bloc soviétique éclate, la région se retrouve séparée du territoire principal de la Russie par deux frontières. Désormais, c’est une enclave russe au milieu de l’Europe de l’Est.Les conflits qui agitent ce lieu encore aujourd’hui sont mis à nu. Tel un champ retourné, il expose un paysage irrégulier, dévoilant simultanément plusieurs couches de son histoire, où les vestiges de l’architecture prussienne et les attributs d’une ville soviétique type se côtoient dans un patchwork aux contrastes improbables. L’éloignement géographique du reste de le Russie, ainsi que son passé hanté par les sujets tabous ont provoqué un délaissement progressif de ce territoire à tous niveaux. Le paysage citadin et rural subissent le même sort, l’abandon.Aujourd’hui Kaliningrad représente au sein de l’Europe une zone qui échappe à la règle, une anomalie, un tiers paysage. Immergé dans un état d’incertitude prolongée quant à son statut et à son devenir, ce territoire évolue suivant ses lois propres, dans l’absence de volonté commune. En l’espace de 70 ans, la frontière dessinée sur une carte d’un territoire uni, s’ancre dans le paysage et devient une scission réelle qui délimite un autre type de civilisation. L’effet de serre qui s’est produit avec l’isolement de ce territoire par rapport à son milieu historique a favorisé l’émergence d’un environnement singulier: passée la frontière, nous avançons dans la réserve de la vieille Europe en friche.Avec Andrei Erofeev, historien d’art de Moscou, nous avons cherché ensemble à comprendre ce phénomène. Il s’agissait pour nous de regarder le paysage de Kaliningrad comme le résultat d’un conflit persistant entre 7 discours – différents types de perception de ce même territoire, qui régissent la relation et le comportement de ses propre habitants. Le sujet nous a ainsi conduits vers une étude multivoque et protéiforme, impliquant notamment un travail d’archives, mais surtout une expérience du territoire réelle, le travail de terrain, nourri de déplacements, d’observations et de rencontres.Mon projet de thèse, K, est issu de ce processus de réflexion et propose une forme de visualisation et d’interprétation plastique de cette recherche. La figure centrale y est celle du terrain vague, empreint successivement de la tentative de table rase du passé européen et du fiasco que connait ici le projet soviétique. C’est un lieu réel et en même temps métaphorique: le cœur de la capitale de la région et à la fois le modèle réduit de l’enclave de Kaliningrad.Les images obtenues via diverses formes d’arpentage du territoire, tendent, d’une part, à en donner une vision d’ensemble, où l’on devine le paysage d’avant, désassemblé. D’autre part, le projet déploie 7 narrations parallèles, constituées d’images et de mots, dédiées au paysage d’après coup.Le projet se partage en deux formes: l’édition et l’exposition, cette dernière étant composée à la fois d’œuvres conçues à partir de la matière documentaire collectée sur place et à distance et passée par un montage, ainsi que de documents bruts. Par ce montage d’éléments trouvés, j’ai essayé d’échapper à la chronologie historique pour proposer un récit qui s’articule autrement – par correspondances et analogies visuelles, par échos et répétitions intertemporelles, par anticipations et rattrapages. L’édition joue le rôle d’introduction sinon d’annexe de l’exposition, le projet se découvre ainsi en deux temps, intervertibles. / This project has never had a research subject as such, but rather an object, an anchor point, a geographical attachment that I chose to gravitate around - Kaliningrad.Formerly part of East Prussia, the region became part of the USSR in 1945 after the end of the Second World War, when this German province was divided between three countries: the USSR, Lithuania and Poland. Its capital, Königsberg, was then renamed Kaliningrad, after Mikhail Kalinin, a collaborator of Stalin. In 1991, when the Soviet bloc broke, the region found itself in the far west of Russia, separated from its mainland by two borders. Henceforth, it is an enclave in the middle of Eastern Europe.The conflicts that agitate this place today are exposed. As a returned field, it shows an uneven landscape, simultaneously revealing several layers of its history - vestiges of the Prussian medieval architecture, and attributes of a typical Soviet town come together in an unlikely patchwork of contrasts. The remoteness of the region from the rest of Russia and it’s past haunted by all sorts of taboos caused gradual abandonment of the territory – cities as well as the countryside suffer the same fate. Today the Kaliningrad region within Europe represents an area exempt from the rules, an anomaly, a "third landscape". Immersed in a prolonged state of uncertainty as to its status and its future, the area evolves according to its own laws, governed by the lack of common will. Thus, in 70 years, the frontier roughly drawn on a map of a united territory, took roots within the landscape and became a real split that delimits another type of civilization. The accidental “greenhouse effect” occurred with the isolation of this territory from its historical environment helped the emergence of a singular landscape: after the border we enter the reserve of the old Europe in decay.Together with Andrei Erofeev, art historian based in Moscow, we tried to understand this phenomenon. We started to consider the landscape of Kaliningrad as the result of a persistent conflict between seven discourses - different types of perception of this same territory that influences the behaviour of its own inhabitants. The theme led us to a multidisciplinary study, involving archival survey, but mostly based on the experience of the territory itself, the fieldwork, guided by displacements, observations and encounters.My thesis, “K”, proposes a form of visualization and plastic interpretation of this research. The central figure in it is the wasteland, imprinted successively by the attempt of tabula rasa of the European past and the fiasco of the Soviet project. The wasteland is a real and also a metaphorical place – it occupies the heart of the capital of the region since several decades and at the same time it represents a reduced model of the entire enclave of Kaliningrad.The images of various forms of surveying the territory tend, on the one hand, to give an overview, where we may guess the previous landscape, disassembled. On the other hand, seven parallel narratives, build with found images and words, express the afterwards landscape. The project is divided in two forms, two phases of reading - the edition and the exhibition, composed of art works based on the montage of documentary material collected on site and remotely, as well as documents shown as such.By assembling found elements I tried to avoid historical chronology and create a narrative articulated differently - through correspondences and visual analogies, anticipations and catch-ups, echoes and cross-temporal repetitions. The edition functions as an introduction or the annex of the exhibition; the project is thus set in two interchangeable stages.

The explosive growth in the number of documents produced daily necessitates the development of effective alternatives to explore, analyze, and discover knowledge from documents. Document mining research work has emerged to devise automated means to discover and analyze useful information from documents. This work has been mainly concerned with constructing text representation models, developing distance measures to estimate similarities between documents, and utilizing that in mining processes such as document clustering, document classification, information retrieval, information filtering, and information extraction. <br /><br /> Conventional text representation methodologies consider documents as bags of words and ignore the meanings and ideas their authors want to convey. It is this deficiency that causes similarity measures to fail to perceive contextual similarity of text passages due to the variation of the words the passages contain, or at least perceive contextually dissimilar text passages as being similar because of the resemblance of words the passages have. <br /><br /> This thesis presents a new paradigm for mining documents by exploiting semantic information of their texts. A formal semantic representation of linguistic inputs is introduced and utilized to build a semantic representation scheme for documents. The representation scheme is constructed through accumulation of syntactic and semantic analysis outputs. A new distance measure is developed to determine the similarities between contents of documents. The measure is based on inexact matching of attributed trees. It involves the computation of all distinct similarity common sub-trees, and can be computed efficiently. It is believed that the proposed representation scheme along with the proposed similarity measure will enable more effective document mining processes. <br /><br /> The proposed techniques to mine documents were implemented as vital components in a mining system. A case study of semantic document clustering is presented to demonstrate the working and the efficacy of the framework. Experimental work is reported, and its results are presented and analyzed.

The explosive growth in the number of documents produced daily necessitates the development of effective alternatives to explore, analyze, and discover knowledge from documents. Document mining research work has emerged to devise automated means to discover and analyze useful information from documents. This work has been mainly concerned with constructing text representation models, developing distance measures to estimate similarities between documents, and utilizing that in mining processes such as document clustering, document classification, information retrieval, information filtering, and information extraction. <br /><br /> Conventional text representation methodologies consider documents as bags of words and ignore the meanings and ideas their authors want to convey. It is this deficiency that causes similarity measures to fail to perceive contextual similarity of text passages due to the variation of the words the passages contain, or at least perceive contextually dissimilar text passages as being similar because of the resemblance of words the passages have. <br /><br /> This thesis presents a new paradigm for mining documents by exploiting semantic information of their texts. A formal semantic representation of linguistic inputs is introduced and utilized to build a semantic representation scheme for documents. The representation scheme is constructed through accumulation of syntactic and semantic analysis outputs. A new distance measure is developed to determine the similarities between contents of documents. The measure is based on inexact matching of attributed trees. It involves the computation of all distinct similarity common sub-trees, and can be computed efficiently. It is believed that the proposed representation scheme along with the proposed similarity measure will enable more effective document mining processes. <br /><br /> The proposed techniques to mine documents were implemented as vital components in a mining system. A case study of semantic document clustering is presented to demonstrate the working and the efficacy of the framework. Experimental work is reported, and its results are presented and analyzed.

Literature digital library is one of the most important resources to preserve civilized asset. To provide more effective and efficient information search, many systems are equipped with a browsing interface that aims to ease the article searching task. A browsing interface is associated with a subject directory, which guides the users to identify articles that need their information need. A subject directory contains a set (or a hierarchy) of subject categories, each containing a number of similar articles. How to group articles in a literature digital library is the theme of this thesis. Previous work used either document classification or document clustering approaches to dispatching articles into a set of article clusters based on their content. We observed that articles that meet a single user¡¦s information need may not necessarily fall in a single cluster. In this thesis, we propose to make use of both Web log and article content is clustering articles. We proposed two hybrid approaches, namely document categorization based method and document clustering based method. These alternatives were compared to other content-based methods. It has been found that the document categorization based method effectively reduces the number of required click-through at the expense of slight increase of entropy that measures the content heterogeneity of each generated cluster.

Effective knowledge management of proliferating volume of documents within a knowledge repository is vital to knowledge sharing, reuse, and assimilation. In order to facilitate accesses to documents in a knowledge repository, use of a knowledge map to organize these documents represents a prevailing approach. Document clustering techniques typically are employed to produce knowledge maps. However, existing document clustering techniques are not tailored to individuals¡¦ preferences and therefore are unable to facilitate the generation of knowledge maps from various preferential perspectives. In response, we propose the Preference-Anchored Document Clustering (PAC) technique that takes a user¡¦s categorization preference (represented as a list of anchoring terms) into consideration to generate a knowledge map (or a set of document clusters) from this specific preferential perspective. Our empirical evaluation results show that our proposed technique outperforms the traditional content-based document clustering technique in the high cluster precision area. Furthermore, benchmarked with Oracle Categorizer, our proposed technique also achieves better clustering effectiveness in the high cluster precision area. Overall, our evaluation results demonstrate the feasibility and potential superiority of the proposed PAC technique.

Knowledge is an increasingly important asset for organizational competition, and knowledge management becomes the most important issue for an organization. Building knowledge ontology is a good solution to increase knowledge reusability. Ontology explicitly defines concepts and their relationships, which can facilitate user understanding and further analysis. Based on previous research (Wu, 2006; Chuang, 2006), this research proposes a refined method for the construction of event ontology. The method includes text pre-processing, event ontology construction, and event ontology presentation. The text pre-processing module includes POS tagger, word filter, and term analysis. Based on the concept of sub-event, we can build a 3-level architecture of event ontology that includes sub-events, events, and topics in the event ontology construction module. Event ontology construction module developed in the project provides a friendly editing environment for the user to edit the concepts and attributes of an event that may cover ¡§who,¡¨ ¡§what,¡¨ ¡§where,¡¨ and ¡§what object.¡¨ In the event ontology presentation module, event episode may be illustrated by event frames, flow charts, and Gantt charts. To verify the feasibility of the proposed method, a prototype system has been built. The Alexander Poison Event was used as an example to demonstrate the value of the prototype system.

The leading effort for creating a standard semantic basis for business documents to solve the electronic business document interoperability problem came from the UN/CEFACT (United Nations Centre for Trade Facilitation and Electronic Business) Core Components Technical Specification (CCTS) through a conceptual document modeling methodology. Currently, the main challenge in using UN/CEFACT CCTS based approaches is that the document artifacts are stored in spreadsheets and this makes it very difficult to discover the previously defined components and to check their consistency. Furthermore, businesses need to customize standard documents according to their specific needs. The first XML implementation of UN/CEFACT CCTS, namely, Universal Business Language (UBL) provides detailed text-based descriptions of customization mechanisms. However, without automated tool support, it is difficult to apply the customization and to maintain the consistency of the customizations. In this thesis, these problems are addressed by providing an online e-business document design and customization environment, i.e. iSURF eDoCreator, which integrates the machine processable versions of paper-based UN/CEFACT CCTTS modeling methodology and UBL customization guidelines, accompanied with an online common UN/CEFACT CCTS based document component repository. In this way, iSURF eDoCreator environment aims to maximize re-use of available document building blocks and minimize the tedious document design and customization efforts. The environment also performs the gap analysis between different customizations of UBL to show how interoperable is the compared document models. The research leading to these results has received funding from the European Community&#039 / s FP7/2007-2013 under grant agreement n&deg / 213031, the iSURF Project.

No description available.

Thesis (Ph. D.)--University of Hong Kong, 2010. / Includes bibliographical references (p. 115-119). Also available in print.

Thesis (M.S. in Computer Science)--Naval Postgraduate School, March 2003. / Thesis advisor(s): Don Brutzman, Curtis L. Blais. Includes bibliographical references (p. 121-122). Also available online.