This thesis gives an account of the nature of entailment through consideration of the role which semantics and syntax have to play in logical theory. This is done in two parts. The first part deals with a class of non-formal logical relations arising from the use of determinate predicates. The second deals with the core of relevance logic and with natural deduction systems of logic and their semantic counterparts. In this part, a philosophical analysis of the concept of normalisability is given, and the significance of transitivity for logic is discussed. The failure of the Lewis proof of B from A & -A is analysed, this analysis giving rise to the notion of suasive power as the key to our understanding of entailment. It is concluded that entailment is to be understood through proper appreciation of its role in clarifying commitments inherent in sets of statements and in particular in revealing contradictions where these occur. For a full account of entailment, both semantically based and syntactically based concepts are necessary.

This thesis investigates logico-philosophical aspects of using either a non-transitive or a non-reflexive logic to obtain a logic of truth in which truth is transparent. It enquires into and rejects the claim that restricting transitivity of entailment to accommodate transparent truth suffices to make the connective tonk acceptable by arguing that tonk as defined in a cut-free sequent calculus requires in addition that the logic is non-reflexive to be uniquely defined, and develops a semantics for tonk based on models with two valuations which delivers a non-transitive and non-reflexive logic. It develops a cut-free sequent calculus and two kinds of semantics for a non-reflexive logic of truth in which truth is transparent, one based on trivalent models and one based on models with two valuations. It shows how to define a non-transitive, a paraconsistent and a paracomplete logic of truth on the models with two valuations and develops a cut-free sequent calculus that captures all four logics. It investigates to which extent the non-reflexive and the non-transitive logic of truth can express their own meta-inferences, and shows among other things how one can employ the paraconsistent and the paracomplete logic to express the meta-inferences of the non-transitive and the non-reflexive logic respectively. Finally, it proves that the non-transitive logic of truth is omega-inconsistent and furthermore that transitivity is not required as assumption to establish that a logic in which truth satisfies the conditions of quantified standard deontic logic is omega-inconsistent.

Measuring semantic similarity is a task needed in many Natural Language Processing (NLP) applications. For example, in Machine Translation evaluation, semantic similarity is used to assess the quality of the machine translation output by measuring the degree of equivalence between a reference translation and the machine translation output. The problem of semantic similarity (Corley and Mihalcea, 2005) is de ned as measuring and recognising semantic relations between two texts. Semantic similarity covers di erent types of semantic relations, mainly bidirectional and directional. This thesis proposes new methods to address the limitations of existing work on both types of semantic relations. Recognising Textual Entailment (RTE) is a directional relation where a text T entails the hypothesis H (entailment pair) if the meaning of H can be inferred from the meaning of T (Dagan and Glickman, 2005; Dagan et al., 2013). Most of the RTE methods rely on machine learning algorithms. de Marne e et al. (2006) propose a multi-stage architecture where a rst stage determines an alignment between the T-H pairs to be followed by an entailment decision stage. A limitation of such approaches is that instead of recognising a non-entailment, an alignment that ts an optimisation criterion will be returned, but the alignment by itself is a poor predictor for iii non-entailment. We propose an RTE method following a multi-stage architecture, where both stages are based on semantic representations. Furthermore, instead of using simple similarity metrics to predict the entailment decision, we use a Markov Logic Network (MLN). The MLN is based on rich relational features extracted from the output of the predicate-argument alignment structures between T-H pairs. This MLN learns to reward pairs with similar predicates and similar arguments, and penalise pairs otherwise. The proposed methods show promising results. A source of errors was found to be the alignment step, which has low coverage. However, we show that when an alignment is found, the relational features improve the nal entailment decision. The task of Semantic Textual Similarity (STS) (Agirre et al., 2012) is de- ned as measuring the degree of bidirectional semantic equivalence between a pair of texts. The STS evaluation campaigns use datasets that consist of pairs of texts from NLP tasks such as Paraphrasing and Machine Translation evaluation. Methods for STS are commonly based on computing similarity metrics between the pair of sentences, where the similarity scores are used as features to train regression algorithms. Existing methods for STS achieve high performances over certain tasks, but poor results over others, particularly on unknown (surprise) tasks. Our solution to alleviate this unbalanced performances is to model STS in the context of Multi-task Learning using Gaussian Processes (MTL-GP) ( Alvarez et al., 2012) and state-of-the-art iv STS features ( Sari c et al., 2012). We show that the MTL-GP outperforms previous work on the same datasets.

Description Logic actions specify adaptations of description logic interpretations based on some preconditions defined using a description logic. We consider DL actions in which preconditions can be specified using DL axioms as well as using conjunctive queries, and combinatiosn thereof. We investigate complexity bounds for the executability and the projection problem for these actions, which respectively ask whether an action can be executed on models of an interpretation, and which entailments are satisfied after an action has been executed on this model. In addition, we consider a set of new reasoning tasks concerned with conflicts and interactions that may arise if two action are executed at the same time. Since these problems have not been investigated before for Description Logic actions, we investigate the complexity of these tasks both for actions with conjunctive queries and without those. Finally, we consider the verification problem for Golog programs formulated over our famility of actions. Our complexity analysis considers several expressive DLs, and we provide tight complexity bounds for those for which the exact complexity of conjunctive query entailment is known.

This dissertation explores the interaction between modality and aspect in Makkan Arabic (MA). There is some consensus in the semantic literature regarding the treatment of modal expressions that may obtain various flavours, such as epistemic, deontic, bouletic, ability, necessity or teleological. These various modal flavours can be captured by a unified lexical entry, and are identified by contextual factors Kratzer (1977, 1981, 1991, 2012). There is some debate regarding the structural location of modal elements, some of which have been argued to be high (the case of epistemic modals) and others low (the case of root modals) (e.g. Cinque (1999)). The relative scope of modals has been subject of much recent work on modality, in particular in relation to their interaction with temporal categories such as aspect. This thesis investigates this topic on the basis of novel data from MA. I observe that the flavour of modality can change depending on how it is inflected with different types of aspect in MA. This observation is in line of Hacquard; Hacquard; Hacquard’s (2006; 2009; 2014) proposal for French and Italian. In MA, when the root modal \gdr\ “can” is inflected with the perfective, the combination yields entailments that have come to be known in the literature as ‘actuality entailments’ (AEs) (see Bhatt (1999, 2006)). In this case, the speaker gives rise to the inference that the proposition expressed by the complement holds in the actual world (instead of merely in some possible but not actual world). My thesis integrates the case of \gdr\ to current cross-linguistic debates on this topic. Building on Hacquard’s work, I argue that AEs are generated when perfective aspect scopes over root modals. Perfective aspect links events to the actual world. Imperfective aspect scoping over the modal fails to generate AEs. My thesis ex- ii tends the investigation of AEs to non-perfective cases. I argue that in addition to the contrast between perfective and imperfective, MA also distinguishes perfect aspect (e.g. an auxiliary plus a modal participle like gaadir). I suggest that the perfect in MA has several shapes, including the choice between two auxiliaries: kaan and saar. I link the different shapes of the perfect to the different types of interpretation identified by Portner (2000, 2003) for the English perfect. I suggest that in MA, different forms of the perfect are linked to distinct interpretations (which in English are grouped together under one form). In addition I show that, contrary to what has been argued by Hacquard for French, the perfect in MA can give rise to AEs in the case of the saar auxiliary. I develop an analysis of the saar perfect that is inspired by Hacquard’s proposal for perfective: in the case of saar, contrary to kaan, the perfect links the eventuality to the actual world. While the discussion of AEs in relation to the modal \gdr\ are linked to the proposal that aspect scopes over the modal, I also examine the case of a modal expression that scopes over aspect: qad “might”. I show that in spite of the fact that aspect scopes below the modal, the contrast between perfective and imperfective in the embedded clause can still give rise to differences in the generation of AEs. This case is interesting because much previous literature on AEs has focused on languages in which aspect scopes over the modal. MA qad provides an example where the modal scopes over aspect, and it is still the case that AEs appear to be generated. In spite of the structural differences with \gdr\, my analysis of qad builds on Hacquard’s proposal for AEs with the perfective, appealing to her proposal for the ‘preservation of event description’ to account for the fact that properties of eventualities can remain stable across worlds. The structure of the thesis is as follows: Chapter 1 provides an introduction to the empirical domain, situating aspect and modality in the description of MA; in addition it iii provides an introduction to key theoretical concepts to be used in later chapters. Chapter 2 discusses AEs in the case of the root modal \gdr\, comparing perfective and imperfective. Chapter 3 extends the discussion of the modal to examples with the perfect, distinguishing between the kaan- and saar- perfects. Chapter 4 investigates the behaviour of qad and its interaction with perfective and imperfective complements. Chapter 5 offers a brief summary and concluding remarks.

A distributed business process is executed in a distributed computing environment. The service-oriented architecture (SOA) paradigm is a popular option for the integration of software services and execution of distributed business processes. Entailment constraints, such as mutual exclusion and binding constraints, are important means to control process execution. Mutually exclusive tasks result from the division of powerful rights and responsibilities to prevent fraud and abuse. In contrast, binding constraints define that a subject who performed one task must also perform the corresponding bound task(s). (authors' abstract)

Avec l'augmentation des connaissances disponibles sur Internet est apparue la difficulté d'obtenir une information. Les moteurs de recherche permettent de retourner des pages Web censés contenir l'information désirée à partir de mots clés. Toutefois il est encore nécessaire de trouver la bonne requête et d'examiner les documents retournés. Les systèmes de questions réponses ont pour but de renvoyer directement une réponse concise à partir d'une question posée en langue naturelle. La réponse est généralement accompagnée d'un passage de texte censé la justifier. Par exemple, pour la question « Quel est le réalisateur d'Avatar ? » la réponse « James Cameron » peut être renvoyée accompagnée de « James Cameron a réalisé Avatar. ». Cette thèse se focalise sur la validation de réponses qui permet de déterminer automatiquement si la réponse est valide. Une réponse est valide si elle est correcte (répond bien à la question) et justifiée par le passage textuel. Cette validation permet d'améliorer les systèmes de questions réponses en ne renvoyant à l'utilisateur que les réponses valides. Les approches permettant de reconnaître les réponses valides peuvent se décomposer en deux grandes catégories : -les approches utilisant un formalisme de représentation particulier de la question et du passage dans lequel les structures sont comparées ;-les approches suivant une approche par apprentissage qui combinent différents critères d'ordres lexicaux ou syntaxiques. Dans le but d'identifier les différents phénomènes sous tendant la validation de réponses, nous avons participé à la création d'un corpus annoté manuellement. Ces phénomènes sont de différentes natures telle que la paraphrase ou la coréférence. On peut aussi remarquer que les différentes informations sont réparties sur plusieurs phrases, voire sont manquantes dans les passages contenant la réponse. Une deuxième étude de corpus de questions a porté sur les différentes informations à vérifier afin de détecter qu'une réponse est valide. Cette étude a montré que les trois phénomènes les plus fréquents sont la vérification du type de la réponse, la date et le lieu contenus dans la question. Ces différentes études ont permis de mettre au point notre système de validation de réponses qui s'appuie sur une combinaison de critères. Certains critères traitent de la présence dans le passage des mots de la question ce qui permet de pointer la présence des informations de la question. Un traitement particulier a été effectué pour les informations de date en détectant une réponse comme n'étant pas valide si le passage ne contient pas la date contenue dans la question. D'autres critères, dont la proximité dans le passage des mots de la question et de la réponse, portent sur le lien entre les différents mots de la question dans le passage. Le second grand type de vérification permet de mesurer la compatibilité entre la réponse et la question. Un certain nombre de questions attendent une réponse étant d'un type particulier. La question de l'exemple précédent attend ainsi un réalisateur en réponse. Si la réponse n'est pas de ce type alors elle est incorrecte. Comme cette information peut ne pas se trouver dans le passage justificatif, elle est recherchée dans des documents autres à l'aide de la structure des pages Wikipédia, en utilisant des patrons syntaxiques ou grâce à des fréquences d'apparitions du type et de la réponse dans des documents. La vérification du type est particulièrement efficace puisqu'elle effectue 80 % de bonnes détections. La vérification de la validité des réponses est également pertinente puisque lors de la participation à une campagne d'évaluation, AVE 2008, le système s'est placé parmi les meilleurs toutes langues confondues. La dernière contribution a consisté à intégrer le module de validation dans un système de questions réponses, QAVAL. Dans ce cadre de nombreuses réponses sont extraites par QAVAL et ordonnées grâce au module de validation de réponses. Le système n'est plus utilisé afin de détecter les réponses valides mais pour fournir un score de confiance à chaque réponse. Le système QAVAL peut ainsi aussi bien être utilisé en effectuant des recherches dans des articles de journaux que dans des articles issus du Web. Les résultats sont assez bons puisqu'ils dépassent ceux obtenus par un simple ordonnancement des réponses de près de 50 %. / Question answering systems extract precise answers from a set of documents, and return the answers along with text snippets which justify them. For example, to the question "Who is the director of Avatar?" The answer "James Cameron" may be returned with "Avatar by James Cameron.".The answer validation detect automatically if the answer is valid ie. if it is correct (responds to the question) and justified by the text passage. This validation allows to improve the question answering systems by producing only valid answers.Two kind of methods can be used to detect right answers : -approaches using specific representation formalism of the question and the passage in which the structures are compared;-learning approaches that combines lexical and syntactic features.To identify the phenomena that characterize the answer validation, we built a manually annotated corpus. Differents phenomena can be seen like paraphrasing, coreference or that the information is spread in different sentences or documents. A second corpus aims to identify the different informations to be checked to valid an answer. This study showed that the three mains phenomena are the answer type, the date and place of the question.These studies have helped to develop our answer validation system which is based on a combination of features. The first one estimates the proportion of common terms in the snippet and the question, the second one measures the proximity of these terms and the answer. The second kind of features measure the compatibility between the answer and the question. Numerous questions wait for answers of an explicit type. For example, the question “Which president succeeded to Jacques Chirac?” requires an instance of president as answer.If the answer is not of this type then it is incorrect. The method aims at verifying that an answer given by a system corresponds to the given type. This verification is done by combining features provided by different methods. The first types of feature are statistical and compute the presence rate of both the answer and the type in documents, other features rely on named entity recognizers and the last criteria are based on the use of Wikipedia. Type checking is particularly effective because it makes 80 % correct detections. The final contribution was to integrate the validation module in a question answering system, QAVAL. Many answers are retrieved by QAVAL and ordered through the answers validation module. The module provide a confidence score to each response. QAVAL can be used both by researching the information in newspaper articles and in articles from the Web. The results are good, exceeding those obtained by a simple answer ranking from nearly 50%.

O reconhecimento de implicação textual (RIT) consiste em identificar automaticamente se um trecho de texto em língua natural é verdadeiro baseado no conteúdo de outro. Este problema vem sendo estudado por pesquisadores da área de Processamento de Línguas Naturais (PLN) há alguns anos, e ganhou certo destaque mais recentemente, com a maior disponibilidade de dados anotados e desenvolvimento de métodos baseados em deep learning. Esta pesquisa de doutorado teve como objetivo o desenvolvimento de recursos e métodos computacionais para o RIT, com especial foco em língua portuguesa. Durante sua realização, foi compilado o corpus ASSIN, o primeiro a fornecer dados para treinamento e avaliação de sistemas de RIT em português, e foi organizado o workshop de mesmo nome, que reuniu pesquisadores interessados no tema. Além disso, foram feitos experimentos computacionais com diferentes tipos de estratégias para o RIT, com dados em inglês e em português. Foi desenvolvido um novo modelo para o RIT, o TEDIN (Tree Edit Distance Network). O modelo é baseado no conceito de distância de edição entre árvores sintáticas, já explorado em outros trabalhos de RIT. Seu diferencial é combinar a representação de conhecimento linguístico explícito com a flexibilidade e capacidade representativa de redes neurais. Foi também desenvolvido o Infernal, um modelo para RIT que usa técnicas clássicas de aprendizado de máquina com engenharia de atributos. Os resultados experimentais do TEDIN ficaram abaixo de outros modelos da literatura, e uma análise cuidadosa de seu comportamento indica a dificuldade de se modelar as diferenças entre árvores sintáticas. Por outro lado, o Infernal teve resultados positivos no ASSIN, definindo o novo estado-da-arte para o RIT em português. / Recognizing Textual Entailment (RTE) consists in automatically identifying whether a text passage in natural language is true based on the content of another one. This problem has been studied in Natural Language Processing (NLP) for some years, and gained some prominence recently, with the availability of annotated data in larger quantities and the development of deep learning methods. This doctoral research had the goal of developing resources and methods for RTE, especially for Portuguese. During its execution, the ASSIN corpus was compiled, which is the first to provide data for training and evaluating RTE systems in Portuguese, and the workshop with the same name was organized, gathering researchers interested in this theme. Moreover, computational experiments were carried out with different techniques for RTE, with English and Portuguese data. A new RTE model, TEDIN (Tree Edit Distance Network), was developed. This model is based on the concept of syntactic tree edit distance, already explored in other RTE works. Its differential is to combine explicit linguistic knowledge representation with the flexibility and representative capacity of neural networks. An RTE model based on classical machine learning and feature engineering, Infernal, was also developed. TEDIN had experimental results below other models from the literature, and a careful analysis of its behavior shows the difficulty of modelling differences between syntactic trees. On the other hand, Infernal had positive results on ASSIN, setting the new stateof- the-art for RTE in Portuguese.

Made available in DSpace on 2016-07-27T13:47:56Z (GMT). No. of bitstreams: 1 Jose Zica dos Santos.pdf: 670352 bytes, checksum: 2ffaf3ab73067afe93dba542f1c63e61 (MD5) Previous issue date: 2005-04-27 / This is an analytical nature dissertation. It is about a research about the phenomenon of the pilgrimage of N. Sra. da Abadia from Dirty Water, this which happens in Minas Gerais. The study seems to be relevant, because, although there is a decline of the Catholic faithfulness participation in the masses and in the sacramental ritual, there is an increasing number of faithful pilgrims at this popular religious type. The main purpose of this work is to reflect about the pilgrim, from the XXI century, his profile and motivation, that built his devotion in the Triângulo Mineiro and the history of this religious fact. The work hypothesis is that the entailment between the devote and saint can be considered as a gift, main category used here to interpret the study. This category is lent by the anthropology, being more specific, by Mauss (1924). There are, analyzing the devotees practices, three elements that take part in this devotion process: the faithful gets something from his protector, a grace, that can be proceeding from the facilitation on getting a material stuff, passing by a body health and also a spiritual health; known and accepted by part of the devotee of this grace, he believes he has the obligation to reattribute to his protector saint something to demonstrate his gratitude. The technique used to get these data is a specific bibliography that studies the origin to this popular piety born in Portugal, transported to Minas Gerais, that perdures and increase up to these days, gathered with a field observation and interviews with the pilgrims in two succeeding years: between August 2003 and August 2004. The main conclusion we got with the dissertation is to evidence that not only the pilgrimages, emitterly popular pieties, can be characterized as a gift, but, in the Brazilian religious field, all the religious manifestations of people can be categorized as a gift. / Esta dissertação é de natureza analítica. Trata-se de uma pesquisa sobre o fenômeno religioso da romaria de N. Sra. da Abadia de Água Suja,festa que ocorre em Minas Gerais. O estudo se mostra relevante, uma vez que, embora haja um declínio da participação dos fiéis católicos nas missas e nos rituais sacramentais, existe cada dia mais um crescente número de fiéis peregrinos neste tipo de religiosidade popular. O objetivo principal do trabalho é refletir sobre o romeiro, ainda do século XXI, seu perfil e sua motivação, que construiu essa devoção no Triângulo Mineiro e a história desse fato religioso. A hipótese de trabalho é a de que o vínculo existente entre o devoto e a Santa pode ser entendido como dom ou dádiva, categoria principal usada aqui para interpretar o estudo. Esta categoria é emprestada pela antropologia, mais precisamente por Mauss (1924). Existem, analisando a prática dos devotos, três elementos que compõem o processo desta devoção: o fiel recebe algo de seu protetor, uma graça, que pode ser proveniente da facilitação na obtenção de um bem material, passando pela saúde corporal e até mesmo espiritual; reconhecida e aceita por parte do devoto esta graça, ele se sente na obrigação de retribuir ao seu Santo protetor algo para demonstrar sua gratidão. A técnica empregada para a obtenção de dados é uma bibliografia específica que historia as origens dessa piedade popular nascida em Portugal, transportada para Minas Gerais, que perdura e cresce até hoje, aliada à observação de campo e entrevistas com os romeiros em dois anos sucessivos: entre agosto de 2003 e agosto de 2004. A principal conclusão que a dissertação chegou é a de constatar que não só as romarias, piedades eminentemente populares, podem ser caracterizadas como dom, mas, no campo religioso brasileiro, todas as manifestações religiosas do povo podem ser categorizadas como dom.

A doença mental é de difícil definição, porque abrange um leque alargado de perturbações que afetam o funcionamento e o comportamento emocional, social e intelectual do indivíduo, mais por desadequação ou distorção do que por falta ou deficiência das suas capacidades anteriores à doença. As doenças mentais manifestam-se em determinado momento, ao longo da vida, antes do qual a pessoa não manifestava alterações ou perda de capacidades. Os objetivos deste estudo são: conhecer o trabalho do profissional-referência desenvolvido junto a pacientes crônicos psiquiátricos, em uma instituição hospitalar; conhecer como o paciente crônico psiquiátrico vê a atuação do profissional-referência, no seu preparo para viver fora do hospital. Pesquisa exploratória que focaliza a atividade do profissional-referência nos cuidados prestados a doentes mentais crônicos. Inicialmente, realizou-se um estudo bibliográfico sobre os temas: reabilitação psicossocial e o trabalho do profissional-referência. Os dados foram coletados em uma instituição apenas, o Hospital Santa Tereza de Ribeirão Preto, hospital psiquiátrico público, fundado no ano de 1944. Centrou-se os dados em um setor específico do hospital, denominado Núcleo de Convívio, onde ficam 18 usuários-moradores. Participaram da pesquisa: todos os profissionais-referência da equipe multidisciplinar que prestam cuidados aos doentes mentais crônicos / institucionalizados, no setor Núcleo de Convívio; todos os clientes, moradores do Núcleo de Convívio, que estavam sob os cuidados da equipe multidisciplinar e dos profissionais-referências da equipe descrita, e que aceitaram participar da pesquisa. Percebemos que apesar do projeto terapêutico desses moradores existir (no papel) e ser discutido com toda a equipe de tempo em tempo (atualizado), ele não aparece na fala nem dos profissionais e nem dos moradores durante a coleta de dados. Não aparecem as atividades propostas para o alcance de objetivos e metas estabelecidas no projeto de cada um dos moradores-referências. Acreditamos que seja de fundamental importância conhecer os gostos, os interesses e as particularidades de cada cliente, para o estabelecimento de um projeto próximo às necessidades de cada pessoa, e neste sentido, as verbalizações demonstram que esse trabalho não está sendo feito pelos profissionais-referências. É necessário uma equipe referência, ou seja, todos profissionais da equipe devem conhecer o projeto de cada morador e assumirem a responsabilidade na execução das implementações. / The mental disease has a difficult definition because it has a wide range of disorders that affect the emotional, social and intellectual behaviour and action of the person, due to disadjustment or distortion than lack or deficience of its former capacities before the disease. The mental sicknesses are expressed in a certain moment during the life and before it the person did not expressed alterations or capacity loss. The aims of this study are: to know the work of a reference professional developed with chronic psychiatric patients in a institution, to know how the chronic psychiatric patient see the reference professional during his preparation to live out of the hospital. From the beginning, it was realized a bibliographic study about the following: psychosocial rehabilitation and the work of the professional reference. These data were collected just in one institution called Hospital Santa Tereza, in Ribeirão Preto- its a public psychiatric hospital, founded in 1944. These data were centered in a specific sector of the hospital, called Núcleo de Convívio, where 18 dwelling patients are settled. Participated in this research all the reference professionals of the multidisciplinary staff that take care of the chronic psychiatric patients/ institutionalized there, all the clients, living at Núcleo de Convívio that were under the responsibility of the multidisciplinary staff and the reference professionals described before, that accepted to participate of this research. We realized that despite of the therapeutic project of this dwelling patients exist (theoretically) and be discussed with all the staff periodically (actualized), it is not presented in the report given by the staff and the patients during the data collection. The proposed activities to the reach of the established aims do not appear in each dwelling patient?s project. We believe that it is a primal importance to know the preferences, interests and peculiarities of each client to the establishment of a project closer to his necessities and, thus, the verbalizations show that this work is not being done by the reference professionals. It Is necessary that a reference staff, that is, all the professionals of the staff must know the each dwelling patient?s project to take over the responsibility in the execution of the improvements.