Spelling suggestions: "subject:"aemantic composition"" "subject:"emantic composition""
1 |
Decoding semantic representations during production of minimal adjective-noun phrasesHonari Jahromi, Maryam 25 April 2019 (has links)
Through linguistic abilities, our brain can comprehend and produce an infinite number of new sentences constructed from a finite set of words. Although recent research has uncovered the neural representation of semantics during comprehension of isolated words or adjective-noun phrases, the neural representation of the words during utterance planning is less understood. We apply existing machine learning methods to Magnetoencephalography (MEG) data recorded during a picture naming
experiment, and predict the semantic properties of uttered words before they are
said. We explore the representation of concepts over time, under controlled tasks,
with varying compositional requirements. Our results imply that there is enough
information in brain activity recorded by MEG to decode the semantic properties of
the words during utterance planning. Also, we observe a gradual improvement in
the semantic decoding of the first uttered word, as the participant is about to say it.
Finally, we show that, compared to non-compositional tasks, planning to compose an
adjective-noun phrase is associated with an enhanced and sustained representation
of the noun. Our results on the neural mechanisms of basic compositional structures
are a small step towards the theory of language in the brain. / Graduate
|
2 |
The Neural Correlates of Basic Semantic Composition: Evidence from fMRI, lesion-behavior mapping and EEGGraessner, Astrid 08 November 2022 (has links)
The ability to combine single words to more complex meanings is the building block of the expressive power of human language. Semantic composition enables us to understand new concepts that we have not encountered before by combining the representations of the underlying individual concepts. The neural correlates of semantic composition have been at the heart of many research agendas in the past but the neural mechanisms at the most basic level still need to be explored consistently. This thesis aimed to advance our knowledge about the spatio-temporal network of brain regions involved in and necessary for basic semantic composition and presents work from three studies applying the same experimental paradigm across three different neuroscientific methodologies. Study I investigated which brain regions support basic semantic composition and how the regions interact via functional magnetic resonance imaging (fMRI). We found differential involvement of left-hemispheric brain regions for a more general combinatorial and a more specific plausibility evaluation process. We furthermore found evidence for an enhanced coupling between two key nodes of the semantic network during successful composition.
In Study II, we probed the relevance of the involved brain regions by studying a cohort of people with aphasia (PWA) and conducting voxel-based lesion-symptom mapping (VLSM). We found a functional dissociation of frontal and temporal brain regions necessary for accurate and efficient semantic decisions, respectively. Finally, in Study III we investigated the temporal dynamics of basic semantic composition in PWA and healthy age-matched controls via electroencephalography (EEG). This study revealed deviant event-related potentials in PWA, where a lack of an early semantic component might be compensated by a stronger late component. Overall, this thesis provides novel insights into the spatial, causal and temporal underpinnings of basic semantic composition making use of three complementary methodologies of neurocognitive research.
|
3 |
Approche stochastique bayésienne de la composition sémantique pour les modules de compréhension automatique de la parole dans les systèmes de dialogue homme-machine / A Bayesian Approach of Semantic Composition for Spoken Language Understanding Modules in Spoken Dialog SystemsMeurs, Marie-Jean 10 December 2009 (has links)
Les systèmes de dialogue homme-machine ont pour objectif de permettre un échange oral efficace et convivial entre un utilisateur humain et un ordinateur. Leurs domaines d'applications sont variés, depuis la gestion d'échanges commerciaux jusqu'au tutorat ou l'aide à la personne. Cependant, les capacités de communication de ces systèmes sont actuellement limités par leur aptitude à comprendre la parole spontanée. Nos travaux s'intéressent au module de compréhension de la parole et présentent une proposition entièrement basée sur des approches stochastiques, permettant l'élaboration d'une hypothèse sémantique complète. Notre démarche s'appuie sur une représentation hiérarchisée du sens d'une phrase à base de frames sémantiques. La première partie du travail a consisté en l'élaboration d'une base de connaissances sémantiques adaptée au domaine du corpus d'expérimentation MEDIA (information touristique et réservation d'hôtel). Nous avons eu recours au formalisme FrameNet pour assurer une généricité maximale à notre représentation sémantique. Le développement d'un système à base de règles et d'inférences logiques nous a ensuite permis d'annoter automatiquement le corpus. La seconde partie concerne l'étude du module de composition sémantique lui-même. En nous appuyant sur une première étape d'interprétation littérale produisant des unités conceptuelles de base (non reliées), nous proposons de générer des fragments sémantiques (sous-arbres) à l'aide de réseaux bayésiens dynamiques. Les fragments sémantiques générés fournissent une représentation sémantique partielle du message de l'utilisateur. Pour parvenir à la représentation sémantique globale complète, nous proposons et évaluons un algorithme de composition d'arbres décliné selon deux variantes. La première est basée sur une heuristique visant à construire un arbre de taille et de poids minimum. La seconde s'appuie sur une méthode de classification à base de séparateurs à vaste marge pour décider des opérations de composition à réaliser. Le module de compréhension construit au cours de ce travail peut être adapté au traitement de tout type de dialogue. Il repose sur une représentation sémantique riche et les modèles utilisés permettent de fournir des listes d'hypothèses sémantiques scorées. Les résultats obtenus sur les données expérimentales confirment la robustesse de l'approche proposée aux données incertaines et son aptitude à produire une représentation sémantique consistante / Spoken dialog systems enable users to interact with computer systems via natural dialogs, as they would with human beings. These systems are deployed into a wide range of application fields from commercial services to tutorial or information services. However, the communication skills of such systems are bounded by their spoken language understanding abilities. Our work focus on the spoken language understanding module which links the automatic speech recognition module and the dialog manager. From the user’s utterance analysis, the spoken language understanding module derives a representation of its semantic content upon which the dialog manager can decide the next best action to perform. The system we propose introduces a stochastic approach based on Dynamic Bayesian Networks (DBNs) for spoken language understanding. DBN-based models allow to infer and then to compose semantic frame-based tree structures from speech transcriptions. First, we developed a semantic knowledge source covering the domain of our experimental corpus (MEDIA, a French corpus for tourism information and hotel booking). The semantic frames were designed according to the FrameNet paradigm and a hand-craft rule-based approach was used to derive the seed annotated training data.Then, to derive automatically the frame meaning representations, we propose a system based on a two decoding step process using DBNs : first basic concepts are derived from the user’s utterance transcriptions, then inferences are made on sequential semantic frame structures, considering all the available previous annotation levels. The inference process extracts all possible sub-trees according to lower level information and composes the hypothesized branches into a single utterance-span tree. The composition step investigates two different algorithms : a heuristic minimizing the size and the weight of the tree ; a context-sensitive decision process based on support vector machines for detecting the relations between the hypothesized frames. This work investigates a stochastic process for generating and composing semantic frames using DBNs. The proposed approach offers a convenient way to automatically derive semantic annotations of speech utterances based on a complete frame hierarchical structure. Experimental results, obtained on the MEDIA dialog corpus, show that the system is able to supply the dialog manager with a rich and thorough representation of the user’s request semantics
|
4 |
Context-aware information systems and their application to health careKawasme, Luay 14 October 2008 (has links)
This thesis explores the field of context-aware information systems (CAIS). We present an approach called Compose, Learn, and Discover (CLD) to incorporate CAIS into the user daily workflow. The CLD approach is self-adjusting. It enables users to personalise the information views for different situations. The CAIS learns about the usage of the information views and recalls the right view in the right situation. We illustrate the CLD approach through an application in the health care field using the Clinical Document Architecture (CDA). In order to realise the CLD approach, we introduce Semantic Composition as a new paradigm to personalise information views. Semantic Composition leverages the type information in the domain model to simplify the user-interface composition process. We also introduce a pattern discovery mechanism that leverages data-mining algorithms to discover correlations between user information needs and different situations.
|
5 |
Composition sémantique pour la langue orale / Semantic composition for spoken language understandingDuvert, Frédéric 10 November 2010 (has links)
La thèse présentée ici a pour but de proposer des systèmes de détection, de composition de constituants sémantiques et d’interprétation dans la compréhension de la langue naturelle parlée. Cette compréhension se base sur un système de reconnaissance automatique de la parole qui traduit les signaux oraux en énoncés utilisables par la machine. Le signal de la parole, ainsi transcrit, comporte un ensemble d’erreurs liées aux erreurs de reconnaissance (bruits, parasites, mauvaise prononciation...). L’interprétation de cet énoncé est d’autant plus difficile qu’il est issu d’un discours parlé, soumis à la disfluence du discours, aux auto-corrections... L’énoncé est de plus agrammatical, car le discours parlé lui-même est agrammatical. L’application de méthodes d’analyses grammaticales ne produit pas de bons résultats d’interprétation, sur des textes issus de transcriptions de la parole. L’utilisation de méthodes d’analyses syntaxiques profondes est à éviter. De ce fait, une analyse superficielle est envisagée. Un des premiers objectifs est de proposer une représentation du sens. Il s’agit de considérer des ontologies afin de conceptualiser le monde que l’on décrit. On peut exprimer les composants sémantiques en logique du premier ordre avec des prédicats. Dans les travaux décrits ici, nous représentons les éléments sémantiques par des frames (FrameNet ). Les structures de frames sont hiérarchisées, et sont des fragments de connaissances auxquels on peut insérer, fusionner ou inférer d’autres fragments de connaissances. Les structures de frames sont dérivables en formules logiques. Nous proposons un système de compréhension de la parole à partir de règles logiques avec le support d’une ontologie, afin de pouvoir créer des liens à partir de composants sémantiques. Puis, nous avons mené une étude sur la découverte des supports syntaxiques des relations sémantiques. Nous proposons une expérience de composition sémantique afin d’enrichir les composants sémantiques de base. Enfin, nous présentons un système de détection de lambda-expression pour mettre en hypothèse les relations à trouver à travers le discours / The thesis presented here is intended to provide detection systems, composition of components and semantic interpretation in the natural spoken language understanding. This understanding is based on an automatic speech recognition system that translates the signals into oral statements used by the machine. The transcribed speech signal, contains a series of errors related to recognition errors (noise, poor pronunciation...). The interpretation of this statement is difficult because it is derived from a spoken discourse, subject to the disfluency of speech, forself-correction... The statement is more ungrammatical, because the spoken discourse itself is ungrammatical. The application of grammatical analysis methods do not produce good results interpretation, on the outcome of speech transcription. The use of deep syntactic analysis methods should be avoided. Thus, a superficial analysis is considered. A primary objective is to provide a representation of meaning. It is considered ontologies to conceptualize the world we describe. We can express the semantic components in first order logic with predicates. In the work described here, we represent the semantic elements by frames (FrameNet ). The frames are hierarchical structures, and are fragments of knowledge which can be inserted, merge or infer other fragments of knowledge. The frames are differentiable structures in logical formulas. We propose a system for speech understanding from logical rules with the support of an ontology in order to create links from semantic components. Then, we conducted a study on the discovery supports syntactic semantic relationships. We propose a compositional semantics experience to enrich the basic semantic components. Finally, we present a detection system for lambda-expression hypothesis to find the relationship through discourse
|
Page generated in 0.1414 seconds