Academic literature on the topic 'Representations contextualisées de mots'

The acquisition and application of knowledge, in particular tacit knowledge (TK), are seen as decisive competitive factors in the knowledge society of the twenty-first century. Despite much talk about the importance of knowledge transfer, little research shows how to identify and measure TK, less research addresses how to transfer TK between individuals and even fewer of these approaches offer any technology that can assist with transfer. This paper does not address the issues of identification and measurement of TK. The purpose of this paper is two-fold: describe a knowledge acquisition and representation technique, known as Ripple Down Rules (RDR), which can be used to capture knowledge, explicit and tacit, in context from those already identified as experts and to describe a set-theoretical technique, known as formal concept analysis (FCA) to assist transfer of the RDR knowledge to another human. Unlike most knowledge acquisition approaches, the RDR knowledge acquisition technique does not rely on the expert to specify what they know. Instead, knowledge becomes codified by the RDR system while the domain expert exercises his or her expertise. The approach does not capture all organisational knowledge, but the knowledge that is captured will be a mixture of different types of knowledge, including formal and codified knowledge that can be learnt from a book and practice-based knowledge that is passed on while on the job. The knowledge captured using RDR may be transferred to another individual through the use of FCA to retrospectively and automatically develop knowledge models that the user can explore. This work offers a possible solution to three knowledge management challenges: capture, utilisation and preservation of knowledge within an organisation.

Purpose The purpose of this paper is to examine the newspaper representations of the aggressive behaviour of social actors in political protests and explore the benefits of integrating corpus linguistics and cognitive approaches to a critical discourse analysis in analysing press reports. Design/methodology/approach This paper uses methods from corpus linguistics and theoretical constructs from cognitive linguistics to examine patterns of representation around Occupy Central, a recent political protest in Hong Kong, in two corpora of English-language newspaper articles published in China Daily and the South China Morning Post (SCMP). Findings An analysis of the ten most frequent collocates of the word police showed that the China Daily corpus articles typically index the presentation of police as vulnerable yet professional in their handling of violent protesters, whereas in SCMP, police officers are often presented as aggressors. The analysis subsequently considered three discursive strategies, namely structural configuration, framing and identification that are mediated through conceptualisations that representations in text evoke. Research limitations/implications In the proposed integrated approach, quantitative investigations of corpus examples could be focussed and contextualised in such a way that particular linguistic instantiations in discourse which are proved statistically salient can be further analysed in relation to conceptual phenomena which serve specific ideological purposes. Originality/value Hopefully, the study could serve as the first ever attempt to adopt an integrative analytical framework in the study of aggression and conflict in news discourse.

This article, the introduction to the special issue “Representations of Women in the French Imaginary: Historicizing the Gallic Singularity,” frames the work of contributors Tracy Adams, Christine Adams, Jean Elisabeth Pedersen, Whitney Walton, and Kathleen Antonioli by analyzing two especially important contemporary debates about French sexual politics, one popular and one academic: (1) the international controversy over Catherine Deneuve’s decision to sign a French manifesto against the American #MeToo movement in Le Monde; and (2) the mixed French and American response to the work of Mona Ozouf in Les mots des femmes: Essai sur la singularité française. The five articles in the special issue itself bring new breadth and depth to the study of these and related debates by exploring a range of different French representations of women in a series of key texts, topics, and historical episodes from the rise of the Middle Ages to the aftermath of World War I.

Systems thinking has been an educational priority for more than a decade, yet its related assessment and teaching strategies have been understudied in the chemistry education research community. Through the lens of systems thinking, this study explores how undergraduate students connect and translate their conceptual representations when they are involved in contextualised problem-solving. The ‘Contextualised Problem Solving’ (CPS) assessment instrument contains four open-ended questions about gas law. Three different cohorts of students registered in a physical science course (2016 Fall, 2017 Spring, 2017 Fall semesters) participated in the problem-solving component of CPS. The results showed that only 8% of students were capable of higher order systems thinking ability when they engaged in problem solving. Over half of the students failed to retrieve essential concepts in problem situations. Most of the participants demonstrated difficulties in organising related systems’ components, understanding the cyclic nature of relationships among systems, and identifying limitations in a specific problem context. By identifying the difficulties and challenges of systems thinking experienced by undergraduate students in solving complex chemistry problems, these findings have the potential to provide fresh insights into effective teaching strategies to promote students’ higher order thinking skills.

BackgroundDespite considerable psychology research being conducted on abortion, there has been no study of the history of psychological knowledge production on the topic. The aim of our research was to analyse journal articles published in English language psychology journals using a politics of location and of representation analytical lens.Study designA systematic search for articles published on abortion in psychology journals from 1960 to 2015 was conducted. A mixed-method approach (content analysis and narrative review) was used to analyse the dataset. Articles were coded according to: decade of publication, region, types of research conducted, and main issues focused on. A narrative review of the dominant issue researched – psychological consequences – in two decades (the 1970s and 2000s) was conducted.ResultsKnowledge production began in the 1970s in most regions featured in the dataset and in the 1990s in South Africa. Research is dominated by quantitative studies conducted in North America and Europe concerning the demarcation of psychological consequences of abortion performed under safe conditions. In the 1970s, abortion was viewed as leading to benign psychological consequences, but by the 2000s traumatology talk was firmly entrenched. Only one article, emanating from South Africa, addressed the question of unsafe abortion.ConclusionsKnowledge production in psychology needs to move beyond a narrow focus on the psychological consequences of abortion and attitudes to abortion. Nuanced, contextualised research of the psychology of both safe and unsafe abortion is necessary.

In the closing phase of the Algerian War in March 1962, Jean-Marie Domenach, director of the journal Esprit, upbraided his counterpart at Les Temps modernes, Jean-Paul Sartre, for failing to understand the greyness of most human actions and the pervasiveness of knots of complicity. Concern for the complexity of complicity was also apparent in Les Temps modernes circles, however, and it was precisely complicity, both in the form of violence of French troops and of the habituation or indifference of the broader French public, that editor Simone de Beauvoir termed a ‘tetanus of the imagination’. Strikingly, she suggested that a means of countering this affliction of getting used to the unconscionable were testimonies of soldiers returning from Algeria in both Les Temps modernes and Esprit. This article examines this mutual concern for the complexities of complicity and investigates its relationship to memory through the curious importance de Beauvoir placed on such testimonies in these two journals. The discussion looks at the mobilisation of the memory of the Second World War in these testimonies, including analogies with fascism and Nazism, and argues that, rather than merely fashionable hyperbole, they powerfully depicted a multifaceted crisis: in Algeria, of French youth, and of France itself. The second part of the article investigates the testimonies’ representation of military institutionalisation – including its detrimental effects on imagination and the facilitation of violence. These representations of systemic or institutional complicity are contextualised alongside scholarly claims that the Algerian war involved a renegotiation of the memory of Vichy France. I argue that the example of these testimonies calls for a qualification of such claims; though they prefigured later conceptions of a complicity memory trope, or ‘the grey zone’ of Vichy France, they did not override the dominant Second World War memory characterised by heroes and victims.

AbstractA film, a television series, four plays and an opera have been produced in China since 1997 dramatising the invasion of Tibet by the British in 1903 04. These works were part of an official effort to enhance patriotic spirit among Chinese and Tibetan people through historical example, as well as an attempt to represent Tibetans as participants in a broader Chinese resistance to Western aggression and humiliation. They coincided with an official call for film-makers to make propaganda more appealing and a decisive turn in Chinese cinema towards commercialised films and Hollywood-style narrative. The paper contextualises these dramatisations and their ideological features within the history of Tibetan representations in Chinese film and television dramas, and discusses foreign critiques of the most influential of the dramatisations of the Younghusband expedition, Feng Xiaonings 1997 film Honghegu (Red River Valley). It notes difficulties with criticisms about the lack of accuracy in these Chinese films, discusses several ways in which they match the historical record, and compares them with the little-known television series Jiangzi 1904.

In this article, I contextualise the emergence and describe the political processes of a grassroots mobilisation against the structural violence of neoliberalism in Mexico in order to suggest the necessity of re-thinking conjunctural analysis in a posthegemonic direction. The National Assembly of the Environmentally Affected (ANAA) is a nationwide network of Mexican communities and organisations that has operated since 2008. ANAA's most notorious achievement has been the opening of a Mexican chapter at the Permanent People's Tribunal, the final verdict of which established the legal responsibility of the Mexican State for structural violence against the Mexican people. My account of ANAA's intervention in the Mexican conjuncture recuperates Stuart Hall's emphasis on complexity and singularity by narrating, through multiple critical voices, the cultural and political conjuncture in which some of the most environmentally affected groups of the Mexican population have been able to organise and strike alliances with critical academic communities or socially concerned scientists. In terms of theoretical elaboration, I reflect on the limits of conjunctural analysis as a response to the deeper crisis of representation – what I call a 'disjuncture' – that concerns the scale of socioenvironmental violence in neoliberal Mexico. In order to think beyond issues of cultural representation, I propose to inform a situated practice of environmentally affected cultural studies with the posthegemonic turn in Latin American thinking of the political.

Abstract. State sequences are a new paradigm to encode and represent contextualised movement data. A state sequence is a temporal succession of characters representing categorical states of the moving entity or its surrounding environment. Eigen decomposition, a principal components analysis method, is an option to reduce and find patterns in such multi-dimensional categorical data through dimensionality reduction. Recurrent patterns can be found by identifying the most relevant eigenbehaviours, which are a set of vectors that characterize the variation in the behaviour of an entity during a time period. Dimensionality reduction techniques have so far not been widely used in movement analytics and in this paper we demonstrate how they could help analyse responses of a moving entity to the dynamic environmental conditions. Specifically, we use sequence-based representation and eigen decomposition to investigate movement patterns of maned wolves (Chrysocyon brachyurus) in relation to vegetation vigour in their habitat. We use a set of GPS-trajectories from a group of maned wolves to which we link multi-source NDVI data as a proxy for the state of vegetation. We find that eigenbehaviours can identify patterns in the wolves’ responses to dynamic environmental conditions that align with the current literature on the species. Our research highlights the potential for dimensionality reduction and sequence-based methods to identify patterns in large tracking databases linked to contextual data.

Resumo: Enquanto existem muitos estudos sobre a reprodução da teoria do “choque das civilizações” e do “mundo islâmico” como o “outro” do Ocidente em mídias ocidentais após os atentados de 11 de setembro de 2001, são extremamente escassos aqueles que analisam a presença dessa teoria nas mídias árabes. Quais são os discursos geopolíticos (re)produzidos nessas mídias com o objetivo de enquadrar e explicar os atentados? Qual é o papel da teoria do “choque das civilizações” e das representações do Ocidente que se manifestam nesse contexto? Este artigo procura responder essas perguntas a partir de uma perspectiva da geopolítica crítica, valendo-se do conceito de “geografias imaginativas”. Os principais jornais transnacionais árabes, al-Hayat, al-Quds al-Arabi e Asharq Al-Awsat, constituem o exemplo empírico. Este artigo mostra que a teoria do “choque das civilizações” – por sua vez duplamente assentada em regionalizações geográficas tradicionais e em uma construção binária entre Ocidente e mundo islâmico – é rejeitada nos jornais analisados. Em vez de se apoiar nessa teoria, os acontecimentos de 11 de setembro de 2001 e os atentados terroristas nos anos seguintes, bem como as geografias imaginativas do Ocidente reproduzidas nesse contexto, são enquadrados e formados por meio de um discurso multifacetado e profundamente pós-colonial. Palavras-chave: Geografias imaginativas. Geopolítica crítica. Mídia impressa árabe. Crítica pós-colonial. NO “CLASH OF CIVILIZATIONS”: AN ANALYSIS OF IMAGINATIVE GEOGRAPHIES IN ARAB PRINT MEDIA AFTER 9/11 Abstract: While many studies analyse the reproduction of the “clash of civilizations” theory and the “Islamic world” as the West’s “other” in Western media after 9/11, there are only few works dedicated to Arab media. What are the geopolitical discourses (re)produced in Arab media framing and explaining the attacks? What is the role of the “clash of civilizations” theory and the representations of the West in this context? This paper addresses these questions from the perspective of critical geopolitics using the concept of “imaginative geographies”. The transnational Arab newspapers al-Hayat, al-Quds al-Arabi e Asharq Al-Awsat serve as the empirical example. I argue that the theory of the “clash of civilizations” – which is based on traditional geographical regionalisations and a binary construction between the West and the Islamic world – is rejected in the analysed newspapers. Instead, 9/11 and other terrorist attacks in the subsequent years as well as the imaginative geographies of the West reproduced in this context are based on a multifaceted and profoundly post-colonial discourse. Keywords: Imaginative Geographies. Critical Geopolitics. Arab Print Media. Postcolonial Critique. AUCUN “CHOC DES CIVILISATIONS”: UNE ANALYSE DES GEOGRAPHIES IMAGINAIRES DANS LES MEDIAS ARABES IMPRIMES APRES LES ATTENTATS DU 11 SEPTEMBRE 2001 Resumé: Bien que de nombreuses études analysent la reproduction de la théorie du “choc des civilisations” et du “monde islamique” comme “l’autre” de l’Occident dans les médias occidentaux après les attentats du 11 septembre 2001, il y a peu de travaux consacrés aux médias arabes. Quels sont les discours géopolitiques (re)produits dans ces médias pour contextualiser et expliquer les attaques ? Quels sont les rôles de la théorie du “choc des civilisations” et de la représentation de l’Occident dans ce contexte ? Cet article tente de répondre à ces questions dans une perspective de géopolitique critique, en utilisant le concept de “géographies imaginaires”. Les principaux journaux transnationaux arabes, al-Hayat, al-Quds al-Arabi e Asharq Al-Awsat sont le matériau de l’enquête. Cet article montre que la théorie du “choc des civilisations” – elle-même basée sur les régionalisations géographiques traditionnelles et la construction opposant Occident et monde islamique – est rejetée dans les journaux analysés. Au lieu de cela, les événements du 11 septembre 2001 comme les attentats terroristes des années suivantes, ainsi que les géographies imaginaires de l’Occident reproduites dans ce contexte, sont façonnés par un discours à multiples facettes, profondément post-colonial. Mots-clés: Géographies Imaginaires. Géopolitique Critique. Médias Imprimés Arabes. Critique Postcoloniale.

Les représentations des mots sont à la base du plupart des systèmes modernes pour le traitement automatique du langage, fournissant des résultats compétitifs. Cependant, d'importantes questions se posent concernant les défis auxquels ils sont confrontés pour faire face aux phénomènes complexes du langage naturel et leur capacité à saisir la variabilité du langage naturel.Pour mieux gérer les phénomènes complexes du langage, de nombreux travaux ont été menées pour affiner les représentations génériques de mots ou pour créer des représentations spécialisées. Bien que cela puisse aider à distinguer la similarité sémantique des autres types de relations sémantiques, il peut ne pas suffire de modéliser certains types de relations, telles que les relations logiques d'implication ou de contradiction.La première partie de la thèse étudie l'encodage de la notion d'implication textuelle dans un espace vectoriel en imposant l'inclusion d'information. Des opérateurs d'implication sont ensuite développées et le cadre proposé peut être utilisé pour réinterpréter un modèle existant de la sémantique distributionnelle. Des évaluations sont fournies sur la détection d'hyponymie en tant que une instance d'implication lexicale.Un autre défi concerne la variabilité du langage naturel et la nécessité de désambiguïser les unités lexicales en fonction du contexte dans lequel elles apparaissent. Les représentations génériques de mots ne réussissent pas à elles seules, des architectures différentes étant généralement utilisées pour aider à la désambiguïsation. Étant donné que les représentations de mots sont construites à partir de statistiques de cooccurrence sur de grands corpus et qu’elles reflètent ces statistiques, elles fournissent une seule représentation pour un mot donné, malgré ses multiples significations. Même dans le cas de mots monosémiques, cela ne fait pas la distinction entre les différentes utilisations d’un mot en fonction de son contexte.Dans ce sens, on pourrait se demander s'il est possible d'exploiter directement les informations linguistiques fournies par le contexte d'un mot pour en ajuster la représentation. Ces informations seraient-elles utiles pour créer une représentation enrichie du mot dans son contexte? Et si oui, des informations de nature syntaxique peuvent-elles aider au processus ou le contexte local suffit? On peux donc examiner si les représentations génériques des mots et la manière dont elles se combinent peut suffire à construire des représentations plus précises.Dans la deuxième partie de la thèse, nous étudions une façon d’incorporer la connaissance contextuelle dans les représentations de mots eux-mêmes, en exploitant les informations provenant de l’analyse de dépendance de phrase ainsi que les informations de voisinage local. Nous proposons des représentations de mots contextualisées sensibles à la syntaxe (SATokE) qui capturent des informations linguistiques spécifiques et encodent la structure de la phrase dans leurs représentations. Cela permet de passer des représentations de type générique (invariant du contexte) à des représentations spécifiques (tenant compte du contexte). Alors que la syntaxe était précédemment considérée pour les représentations de mots, ses avantages n'ont peut-être pas été entièrement évalués au-delà des modèles qui exploitent ces informations à partir de grands corpus.Les représentations obtenues sont évaluées sur des tâches de compréhension du langage naturel: classification des sentiments, détection de paraphrases, implication textuelle et analyse du discours. Nous démontrons empiriquement la supériorité de ces représentations par rapport aux représentations génériques et contextualisées des mots existantes.Le travail proposé dans la présente thèse contribue à la recherche dans le domaine de la modélisation de phénomènes complexes tels que l'implication textuelle, ainsi que de la variabilité du langage par le biais de la proposition de représentations contextualisés
Continuous word representations (word type embeddings) are at the basis of most modern natural language processing systems, providing competitive results particularly when input to deep learning models. However, important questions are raised concerning the challenges they face in dealing with the complex natural language phenomena and regarding their ability to capture natural language variability.To better handle complex language phenomena, much work investigated fine-tuning the generic word type embeddings or creating specialized embeddings that satisfy particular linguistic constraints. While this can help distinguish semantic similarity from other types of semantic relatedness, it may not suffice to model certain types of relations between texts such as the logical relations of entailment or contradiction.The first part of the thesis investigates encoding the notion of entailment within a vector space by enforcing information inclusion, using an approximation to logical entailment of binary vectors. We further develop entailment operators and show how the proposed framework can be used to reinterpret an existing distributional semantic model. Evaluations are provided on hyponymy detection as an instance of lexical entailment.Another challenge concerns the variability of natural language and the necessity to disambiguate the meaning of lexical units depending on the context they appear in. For this, generic word type embeddings fall short of being successful by themselves, with different architectures being typically employed on top to help the disambiguation. As type embeddings are constructed from and reflect co-occurrence statistics over large corpora, they provide one single representation for a given word, regardless of its potentially numerous meanings. Furthermore, even given monosemous words, type embeddings do not distinguish between the different usages of a word depending on its context.In that sense, one could question if it is possible to directly leverage available linguistic information provided by the context of a word to adjust its representation. Would such information be of use to create an enriched representation of the word in its context? And if so, can information of syntactic nature aid in the process or is local context sufficient? One could thus investigate whether looking at the representations of the words within a sentence and the way they combine with each-other can suffice to build more accurate token representations for that sentence and thus facilitate performance gains on natural language understanding tasks.In the second part of the thesis, we investigate one possible way to incorporate contextual knowledge into the word representations themselves, leveraging information from the sentence dependency parse along with local vicinity information. We propose syntax-aware token embeddings (SATokE) that capture specific linguistic information, encoding the structure of the sentence from a dependency point of view in their representations. This enables moving from generic type embeddings (context-invariant) to specific token embeddings (context-aware). While syntax was previously considered for building type representations, its benefits may have not been fully assessed beyond models that harvest such syntactical information from large corpora.The obtained token representations are evaluated on natural language understanding tasks typically considered in the literature: sentiment classification, paraphrase detection, textual entailment and discourse analysis. We empirically demonstrate the superiority of the token representations compared to popular distributional representations of words and to other token embeddings proposed in the literature.The work proposed in the current thesis aims at contributing to research in the space of modelling complex phenomena such as entailment as well as tackling language variability through the proposal of contextualized token embeddings

Depuis une vingtaine d'annees, des relocalisations territoriales ainsi qu'une redefinition du pouvoir local sont en jeu a nantes comme dans la plupart des grandes villes francaises. On note egalement une evolution dans la facon de cerner les problemes et les enjeux de la ville. Ces territorialisations des enjeux urbains passent par les differentes representations de la ville et se traduisent notamment par la maniere de nommer et de caracteriser les espaces auxquels on se refere, ce que j'appelle la mise en mots de la ville contemporaine. Plus que la ville de nantes, l'objet de cette recherche porte sur ses representations, car si la ville ne se laisse pas saisir hors de sa materialite, elle n'est produite comme unite qu'au travers du sens qu'on lui attribue. Il convient donc de decrypter les discours sur nantes et d'analyser comment, a partir de ces discours, se degagent des << representations en actes >>. Dans l'hypothese que l'emergence des images urbaines est consubstantielle a la construction de projets territoriaux portes par de nouveaux acteurs urbains, ce sont les rapports dialectiques entre les representations et les systemes d'acteurs locaux qui sont au centre de cet exercice universitaire. En partant du principe que chaque acteur, inscrit dans un groupe social, a une vision, un angle d'attaque pour apprehender la ville mais egalement des objectifs et une pratique specifiques au role et a la place qu'il occupe, il est interessant de confronter les differentes images constitutives d'une representation, ainsi que de degager l'existence d'ideologies urbaines et de strategies territoriales

Cette thèse aborde les mots "cuico" et "flaite" comme des catégories qui sont devenues fondamentales dans la représentation des classes sociales aujourd’hui au Chili. Fruits à la fois de transformations sociales et culturelles récentes, et héritières de rapports historiques qui ont été tissés entre les différents groupes qui composent notre société, ces deux catégories expriment dans un langage courant et informel la méfiance qu’il existe de nos jours entre les classes sociales dans le pays. Constituant toutes les deux des catégories péjoratives, cuico, associée à la classe supérieure, et flaite, associée aux classes populaires dangereuses, servent non seulement à identifier des groupes reconnus comme différents et spécifiques en ce moment dans la société chilienne, mais aussi à tracer des frontières morales entre les groupes. Elles rendent compte ainsi de l’importance du langage moral lorsqu’on parle des classes sociales aujourd’hui au Chili
This thesis addresses the terms cuico and flaite as categories that have become fundamental in the representation of social classes in Chile today. They are the result of the recent social and cultural transformations and the historical reports that have been woven between groups that compose our society. These two categories express in a language both common and informal the mistrust that exists these days between social classes. Both constituting pejorative categories, cuico, associated to the upper class, and flaite, associated with dangerous popular classes, serve not only to identify this different and specific groups in this moment in Chilean society, but also to trace moral boundaries between the two. They account for the importance of moral language when speaking of social classes in Chile today

Dans l’enseignement des langues étrangères en Colombie, l’intégration de la culture aux cours de FLE tourne autour de trois concepts clé : civilisation, culture et interculturel. Pour pouvoir comprendre les pratiques enseignantes et la forte tendance à limiter l’enseignement de la culture à des caractéristiques folklorisantes, il serait nécessaire de comprendre la représentation linguistique que les enseignants ont de chacun de ces concepts, ainsi que la construction identitaire qu’ils véhiculent. Seul une approche sémantique complétée avec une analyse du rapport au savoir, pourrait nous permettre de comprendre les pratiques didactiques effectives dans la salle de classe. L’étude que nous présentons, s’inscrit donc dans le croissement de la Sémantique des Possibles Argumentatifs et la didactique des langues étrangères, à travers la mise en place d’un protocole expérimental intégrant une approche sémantique à l’analyse du rapport au savoir. Pour avoir accès aux acteurs directs de l’enseignement de français dans le secteur public en Colombie, nous avons décidé d’inscrire notre étude dans le cadre du Projet de réintroduction de français, que le Ministère de l’Éducation Nationale et l’Ambassade de France ont entrepris depuis 2009 dans le pays. Ainsi, notre étude compte sur la participation de huit enseignants et de trente-neuf futurs enseignants de français. Nous croyons avoir démontré que le protocole expérimental mis en place, associant le protocole de la SPA, des bilans de savoirs et de l’entretien d’explicitation permet l’analyse de la singularité de la pratique enseignante. Cette approche s’avère plus compréhensive des actes pédagogiques, mais souligne également les obstacles provoqués par le manque de formation en méthodologie et didactique des langues étrangères. En réalité, les choix pédagogiques mettent en jeu la représentation sociale de « soi » et de l’autre » car ils véhiculent une certaine construction identitaire et donc, outre la bonne volonté des enseignants, ces choix devraient être plus conscients
The integration of culture in FFL (French as a Foreign Language) teaching in Colombia deals with three key concepts: Civilisation, culture and interculturality. In order to understand teaching practices and their limitation to teach culture based on folkloric images of people, it will be necessary to examine teachers’ linguistic representations of these three concepts as well as the identity construction that teachers’ representations spread. Only; a semantic approach accompanied by a relation to knowledge analysis, would allow us to understand the actual teaching practices in the classroom. This study is placed at the intersection between the Semantic of Argumentative Possibilities and foreign languages didactics, thanks to the implantation of an experimental protocol that incorporates a semantic approach into the analyses obtained through relation to knowledge In order to be directly in contact with French teachers from the state schools of Colombia, we have chosen the reintroduction of French Teaching project (Proyecto de reintroduccion del francés) as the context for this study. This project started in 2009 thanks to the efforts of the Ministry of National Education and the French embassy in Colombia. The participants were divided into two groups. One of 8 French teachers and the other of 39 future French teachers, who were in training and responsible for the teaching of French in primary and secondary schools. We believe that the experimental protocol used in this research is a way to analyse the singularity of teachers’ practices: This approach seems to provide a good understanding of teaching practices, even if it underlines problems caused by the lack of training in the field of methodology and didactics of foreign languages. As a matter of fact, the pedagogical choices put the social representation of “the self” and “the other” at stake, because they spread a certain identity construction and despite of the teachers’ good will, teachers should be aware of their choices

De nombreuses applications en traitement du langage naturel (TALN) reposent sur les représentations de mots, ou “word embeddings”. Ces représentations doivent capturer à la fois de l’information syntaxique et sémantique pour donner des bonnes performances dans les tâches en aval qui les utilisent. Cependant, les méthodes courantes pour les apprendre utilisent des textes génériques comme Wikipédia qui ne contiennent pas d’information sémantique précise. De plus, un espace mémoire important est requis pour pouvoir les sauvegarder car le nombre de représentations de mots à apprendre peut être de l’ordre du million. Le sujet de ma thèse est de développer de nouveaux algorithmes pour améliorer l’information sémantique dans les word embeddings tout en réduisant leur taille en mémoire lors de leur utilisation dans des tâches en aval de TALN.La première partie de mes travaux améliore l’information sémantique contenue dans les word embeddings. J’ai développé dict2vec, un modèle qui utilise l’information des dictionnaires linguistiques lors de l’apprentissage des word embeddings. Les word embeddings appris par dict2vec obtiennent des scores supérieurs d’environ 15% par rapport à ceux appris avec d’autres méthodes sur des tâches de similarités sémantiques de mots. La seconde partie de mes travaux consiste à réduire la taille mémoire des word embeddings. J’ai développé une architecture basée sur un auto-encodeur pour transformer des word embeddings à valeurs réelles en vecteurs binaires, réduisant leur taille mémoire de 97% avec seulement une baisse de précision d’environ 2% dans des tâches de TALN en aval
Many natural language processing applications rely on word representations (also called word embeddings) to achieve state-of-the-art results. These numerical representations of the language should encode both syntactic and semantic information to perform well in downstream tasks. However, common models (word2vec, GloVe) use generic corpus like Wikipedia to learn them and they therefore lack specific semantic information. Moreover it requires a large memory space to store them because the number of representations to save can be in the order of a million. The topic of my thesis is to develop new learning algorithms to both improve the semantic information encoded within the representations while making them requiring less memory space for storage and their application in NLP tasks.The first part of my work is to improve the semantic information contained in word embeddings. I developed dict2vec, a model that uses additional information from online lexical dictionaries when learning word representations. The dict2vec word embeddings perform ∼15% better against the embeddings learned by other models on word semantic similarity tasks. The second part of my work is to reduce the memory size of the embeddings. I developed an architecture based on an autoencoder to transform commonly used real-valued embeddings into binary embeddings, reducing their size in memory by 97% with only a loss of ∼2% in accuracy in downstream NLP tasks

Dans cette thèse, nous nous concentrons sur comment les représentations parcimonieuses peuvent aider à augmenter les performances pour réduire le bruit, extraire des régions de texte, reconnaissance des formes et localiser des symboles dans des documents graphiques. Pour ce faire, tout d'abord, nous donnons une synthèse des représentations parcimonieuses et ses applications en traitement d'images. Ensuite, nous présentons notre motivation pour l'utilisation de dictionnaires d'apprentissage avec des algorithmes efficaces pour les construire. Après avoir décrit l'idée générale des représentations parcimonieuses et du dictionnaire d'apprentissage, nous présentons nos contributions dans le domaine de la reconnaissance de symboles et du traitement des documents en les comparants aux travaux de l'état de l'art. Ces contributions s'emploient à répondre aux questions suivantes: La première question est comment nous pouvons supprimer le bruit des images où il n'existe aucune hypothèse sur le modèle de bruit sous-jacent à ces images ? La deuxième question est comment les représentations parcimonieuses sur le dictionnaire d'apprentissage peuvent être adaptées pour séparer le texte du graphique dans des documents? La troisième question est comment nous pouvons appliquer la représentation parcimonieuse à reconnaissance de symboles? Nous complétons cette thèse en proposant une approche de localisation de symboles dans les documents graphiques qui utilise les représentations parcimonieuses pour coder un vocabulaire visuel
In this thesis, we focus on how sparse representations can help to increase the performance of noise removal, text region extraction, pattern recognition and spotting symbols in graphical documents. To do that, first of all, we give a survey of sparse representations and its applications in image processing. Then, we present the motivation of building learning dictionary and efficient algorithms for constructing a learning dictionary. After describing the general idea of sparse representations and learned dictionary, we bring some contributions in the field of symbol recognition and document processing that achieve better performances compared to the state-of-the-art. These contributions begin by finding the answers to the following questions. The first question is how we can remove the noise of a document when we have no assumptions about the model of noise found in these images? The second question is how sparse representations over learned dictionary can separate the text/graphic parts in the graphical document? The third question is how we can apply the sparse representation for symbol recognition? We complete this thesis by proposing an approach of spotting symbols that use sparse representations for the coding of a visual vocabulary

L’émergence, dès la seconde moitié des années 1950, de deux milieux de la critique architecturale à Berlin-est et à Berlin-ouest, révèle, malgré les contraintes politiques qui pèsent sur ses acteurs, une volonté partagée de réflexion sur l’identité urbaine de la ville divisée. En effet, la Reconstruction des villes allemandes après la Seconde Guerre mondiale désigne un vaste projet de remise en service des infrastructures matérielles, mais aussi, de redéfinition d’un champ culturel et intellectuel profondément marqué par le nazisme. Adoptant une perspective d’histoire culturelle et des intellectuels, cette thèse explore les carrières, écrits et réseaux d’Ulrich Conrads, Günther Kühne, Julius Posener, Wolf Jobst Siedler, Klaus Duntze, Dieter Hoffmann-Axthelm et Nikolaus Kuhnert à Berlin-ouest, et de Kurt Liebknecht, Richard Paulick, Hermann Henselmann, Kurt Junghanns, Bruno Flierl et Wolfgang Kil à Berlin-est. Architectes ou historiens de l’art de formation, auteurs dans la presse spécialisée, et journalistes dans la presse généraliste écrite et radiophonique, mais aussi, parfois, éditeurs et professeurs dans les universités et écoles d’art, ces critiques d’architecture et de la ville s’engagèrent, entre les années 1950 et 1980, pour la reconnaissance d’une culture architecturale [Baukultur] berlinoise au XX ͤ siècle. Ils furent aussi, à bien des égards, les initiateurs d’une redéfinition des enjeux de la critique architecturale et urbaine dans l’Allemagne d’après 1945
Both in East and West Berlin, as soon as the mid-fifties, architecture critics have come to the fore and shown that in spite of political pressures imposed on them, they shared a will to reflect upon the urban identity of the divided city. Indeed, the Reconstruction of German towns after WWII refers to a vast project of restoration of the material infrastructures but also the redefinition of a cultural and intellectual landscape bearing the heavy mark of Nazi ideology. Choosing a perspective of cultural and intellectual history, this thesis explores the careers, writings and networks of Ulrich Conrads, Günther Kühne, Julius Posener, Wolf Jobst Siedler, Klaus Duntze, Dieter Hoffmann-Axthelm and Nikolaus Kunhert in West-Berlin and of Kurt Liebknecht, Richard Paulick, Hermann Henselmann, Kurt Junghanns, Bruno Flierl and Wolfgang Kil in East-Berlin. Whether architects or history of art specialists, writers in the specialist press, journalists in the general press but also sometimes publishers and teachers in universities and art schools, those critics committed themselves to making the Berlin architectural culture [Baukultur] known in the XXth century. To a large extent they initiated a redefining of the stakes of architectural and urban criticism in post 1945 Germany

Image representation is in the heart of many computer vision algorithms. Different computer vision tasks (e.g. classification, detection) require discriminative image representations to recognize visual categories. In a nutshell, the bag-of-visual-words image representation is the most successful approach for object and scene recognition. In this thesis, we mainly revolve around this model and search for discriminative image representations. In the first part, we present a novel approach to incorporate spatial information in the BoVW method. In this framework, we present a simple and efficient way to infuse spatial information by taking advantage of the orientation and length of the segments formed by pairs of similar descriptors. We introduce the notion of soft-similarity to compute intra and inter visual word spatial relationships. We show experimentally that, our method adds important discriminative information to the BoVW method and complementary to the state-of-the-art method. Next, we focus on color description in general. Differing from traditional approaches of invariant description to account for photometric changes, we propose discriminative color descriptor. We demonstrate that such a color description automatically learns a certain degree of photometric invariance. Experiments show that the proposed descriptor outperforms existing photometric invariants. Furthermore, we show that combined with shape descriptor, the proposed color descriptor obtain excellent results on four challenging data sets.Finally, we focus on the most accurate color representation i.e. multispectral reflectance which is an intrinsic property of a surface. Even with the modern era technological advancement, it is difficult to extract reflectance information without sophisticated instruments. To this end, we propose to use the display of the device as an illuminant while the camera captures images illuminated by the red, green and blue primaries of the display. Three illuminants and three response functions of the camera lead to nine response values which are used for reflectance estimation. Results show that the accuracy of the spectral reconstruction improves significantly over the spectral reconstruction based on a single illuminant. We conclude that, multispectral data acquisition is potentially possible with consumer hand-held devices such as tablets, mobiles, and laptops

Nous abordons, dans cette thèse, une étude sur les représentations continues de mots (en anglais word embeddings) appliquées à la détection automatique des erreurs dans les transcriptions de la parole. Notre étude se concentre sur l’utilisation d’une approche neuronale pour améliorer la détection automatique des erreurs dans les transcriptions automatiques, en exploitant les word embeddings. L’exploitation des embeddings repose sur l’idée que la détection d’erreurs consiste à trouver les possibles incongruités linguistiques ou acoustiques au sein des transcriptions automatiques. L’intérêt est donc de trouver la représentation appropriée du mot qui permet de capturer des informations pertinentes pour pouvoir détecter ces anomalies. Notre contribution dans le cadre de cette thèse porte sur plusieurs axes. D’abord, nous commençons par une étude préliminaire dans laquelle nous proposons une architecture neuronale capable d’intégrer différents types de descripteurs, y compris les embeddings. Ensuite, nous nous focalisons sur une étude approfondie des représentations continues de mots. Cette étude porte d’une part sur l’évaluation de différents types d’embeddings linguistiques puis sur leurs combinaisons. D’autre part, elle s’intéresse aux embeddings acoustiques de mots. Puis, nous présentons une étude sur l’analyse des erreurs de classifications, qui a pour objectif de percevoir les erreurs difficiles à détecter.Finalement, nous exploitons les embeddings linguistiques et acoustiques ainsi que l’information fournie par notre système de détections d’erreurs dans plusieurs cadres applicatifs
My thesis concerns a study of continuous word representations applied to the automatic detection of speech recognition errors. Our study focuses on the use of a neural approach to improve ASR errors detection, using word embeddings. The exploitation of continuous word representations is motivated by the fact that ASR error detection consists on locating the possible linguistic or acoustic incongruities in automatic transcriptions. The aim is therefore to find the appropriate word representation which makes it possible to capture pertinent information in order to be able to detect these anomalies. Our contribution in this thesis concerns several initiatives. First, we start with a preliminary study in which we propose a neural architecture able to integrate different types of features, including word embeddings. Second, we propose a deep study of continuous word representations. This study focuses on the evaluation of different types of linguistic word embeddings and their combination in order to take advantage of their complementarities. On the other hand, it focuses on acoustic word embeddings. Then, we present a study on the analysis of classification errors, with the aim of perceiving the errors that are difficult to detect. Perspectives for improving the performance of our system are also proposed, by modeling the errors at the sentence level. Finally, we exploit the linguistic and acoustic embeddings as well as the information provided by our ASR error detection system in several downstream applications

La recherche a pour objet la structuration du champ lexical des mots « chance », « fortune », « hasard » et « risque » du XVIIIe au XXIe siècle. Témoin de mutations qu’a connues la civilisation occidentale durant cette période, ce champ, qui se rattache à la notion de fortune / hasard, présente une relative homogénéité sémantique.Les mots (et leurs dérivés) sont étudiés à travers le déploiement, la régulation et la répartition des normes d’usages, non seulement en français hexagonal, mais aussi en français québécois. L’étude se fonde sur l’exploitation de deux types de corpus. D’une part, un corpus d’articles extraits d’une cinquantaine de dictionnaires sert à mettre en évidence la productivité morphosémantique et sémantique de ces unités dans une perspective historique large. D’autre part, un grand ensemble d’énoncés diversifiés permet, par la mise au jour de types de contextes, d’effectuer un suivi diachronique des usages. L’approche continuiste des différences d’usages s’appuie sur une représentation fréquentielle des changements sémantiques.La thèse apporte une contribution à la question de la variation des usages et du changement sémantique, qui ouvre sur plusieurs perspectives. Elle se veut d’abord une réflexion sur la théorie et la méthodologie descriptives, appréhendées à la lumière de l’analyse de la nature et du rôle des corpus. Elle met ensuite en évidence l’importance de la dimension intersubjective dans l’activité de signification, en particulier le rôle déterminant des structures syntagmatiques dans l’établissement de nouveaux usages sémantiques. Enfin, elle permet de mettre en relation le changement sémantique avec les conditions sociohistoriques et les représentations collectives
The present study deals with the way in which the lexical field regrouping the words “chance”, “fortune”, “hazard” and “risqué” has been structured in the French language from the eighteenth century till the present day. Revealing major changes in western societies during this period of time, the field, which corresponds to the linguistic representation of the notion of fortune / hasard, presents a certain coherence.We have examined these words and their derived forms through the display, regulation, and distribution of norms of use, not only in Hexagonal French, but also in Quebec French. Two types of corpora have been analysed. On the one hand, a corpus of articles from around 50 dictionaries has been used to emphasize the lexical and semantic productivity of the different units on a large historical scale. On the other hand, in revealing context types, a set of texts reflecting French language varieties has allowed for carrying out a diachronic analysis of lexical uses. The continuist approach to semantic differences rests upon a frequential representation of semantic changes.The thesis brings a significant contribution to the question of usage variations and semantic change, providing new perspectives. It first deals with theory and methodology of lexical description, considered through the analysis of the nature and the role of corpora. It then evidences the central role of syntagmatic structures in the setting of new semantic uses. The study has finally put into relation semantic changes with their historical background and the collective representations of the time

Résumé: Les représentations des villes depuis la mer ont eu, à partir du XVe siècle, un grand succès dans l’iconographie urbaine des voyageurs. Depuis 1911, Le Corbusier a pris l’habitude d’esquisser dans ses carnets de croquis les paysages urbains découverts au cours de son voyage en Orient. Villes, remparts ou monuments significatifs sont souvent représentés depuis la mer ou les fleuves, car l'arrivée en bateau autorise une perception progressive et globale du site. Suite à son séjour en Amérique du sud en 1929, l'architecte exprime ouvertement son appréciation pour la perspective obtenue depuis le bateau au milieu des baies, car ce point de vue lui permet de contextualiser immédiatement ses propositions d’aménagement urbain. De simple connaissance des villes, cette pratique iconographique devient, chez Le Corbusier, un instrument d’élaboration du projet. Mais la vue horizontale est bientôt accompagnée du procédé de survol des centres urbains pour appréhender la grandeur du paysage naturel et construit. À travers l’analyse des nombreuses images – dessins, croquis, photos – élaborées par Le Corbusier à partir des différents moyens de transport, ainsi que des commentaires accompagnant ces documents iconographiques et de quelques textes publiés, on essayera de mieux comprendre le rapport entre l’échelle de ces visions panoramiques et la nouvelle approche territoriale mise au point par Le Corbusier à partir des années Trente. Abstract: The representations of cities viewed from the sea have had, since the fifteenth century, a great success in urban iconography. From 1911, Le Corbusier began drawing urban landscapes in his sketchbooks, which he discovered during his trip to the Orient. Cities, fortifications or significant monuments are often depicted from the sea or rivers, because the arrival by boat allows a gradual and overall perception of the site. After his stay in South America in 1929, the architect openly expressed his appreciation for the panoramic view obtained from the boat in the middle of the bay. This point of view allows him to contextualize his urban development proposals immediately. This iconographic practice is no longer a simple function of knowledge, but it has become an instrument for realising the project. However, the horizontal view was soon accompanied by the process of flying over urban centres to understand the magnitude of the man-made and natural landscape. Through the analysis of the many images – drawings, sketches, photos – developed by Le Corbusier using different means of transport, of comments accompanying these iconographic documents and of some published texts, the paper aims to better understand the relationship between the scale of these panoramic visions and the new territorial approach developed by Le Corbusier from the Thirties onwards. Mots clés: moyens de transport; avion; bateau. Keywords: means of transport; airplane; boat DOI: http://dx.doi.org/10.4995/LC2015.2015.821