Statistical shift-reduce parsing involves the interplay of representation learning, structured learning, and inexact search. This dissertation considers approaches that tightly integrate these three elements and explores three novel models for shift-reduce CCG parsing. First, I develop a dependency model, in which the selection of shift-reduce action sequences producing a dependency structure is treated as a hidden variable; the key components of the model are a dependency oracle and a learning algorithm that integrates the dependency oracle, the structured perceptron, and beam search. Second, I present expected F-measure training and show how to derive a globally normalized RNN model, in which beam search is naturally incorporated and used in conjunction with the objective to learn shift-reduce action sequences optimized for the final evaluation metric. Finally, I describe an LSTM model that is able to construct parser state representations incrementally by following the shift-reduce syntactic derivation process; I show expected F-measure training, which is agnostic to the underlying neural network, can be applied in this setting to obtain globally normalized greedy and beam-search LSTM shift-reduce parsers.

linguagem é uma marca da humanidade e da consciência, sendo a conversação (ou diálogo) uma das maneiras de comunicacão mais fundamentais que aprendemos quando crianças. Por isso uma forma de fazer um computador mais atrativo para interação com usuários é usando linguagem natural. Dos sistemas com algum grau de capacidade de linguagem desenvolvidos, o chatterbot Eliza é, provavelmente, o primeiro sistema com foco em diálogo. Com o objetivo de tornar a interação mais interessante e útil para o usuário há outras aplicações alem de chatterbots, como agentes conversacionais. Estes agentes geralmente possuem, em algum grau, propriedades como: corpo (com estados cognitivos, incluindo crenças, desejos e intenções ou objetivos); incorporação interativa no mundo real ou virtual (incluindo percepções de eventos, comunicação, habilidade de manipular o mundo e comunicar com outros agentes); e comportamento similar ao humano (incluindo habilidades afetivas). Este tipo de agente tem sido chamado de diversos nomes como agentes animados ou agentes conversacionais incorporados. Um sistema de diálogo possui seis componentes básicos. (1) O componente de reconhecimento de fala que é responsável por traduzir a fala do usuário em texto. (2) O componente de entendimento de linguagem natural que produz uma representação semântica adequada para diálogos, normalmente utilizando gramáticas e ontologias. (3) O gerenciador de tarefa que escolhe os conceitos a serem expressos ao usuário. (4) O componente de geração de linguagem natural que define como expressar estes conceitos em palavras. (5) O gerenciador de diálogo controla a estrutura do diálogo. (6) O sintetizador de voz é responsável por traduzir a resposta do agente em fala. No entanto, não há consenso sobre os recursos necessários para desenvolver agentes conversacionais e a dificuldade envolvida nisso (especialmente em línguas com poucos recursos disponíveis). Este trabalho foca na influência dos componentes de linguagem natural (entendimento e gerência de diálogo) e analisa em especial o uso de sistemas de análise sintática (parser) como parte do desenvolvimento de agentes conversacionais com habilidades de linguagem mais flexível. Este trabalho analisa quais os recursos do analisador sintático contribuem para agentes conversacionais e aborda como os desenvolver, tendo como língua alvo o português (uma língua com poucos recursos disponíveis). Para isto, analisamos as abordagens de entendimento de linguagem natural e identificamos as abordagens de análise sintática que oferecem um bom desempenho. Baseados nesta análise, desenvolvemos um protótipo para avaliar o impacto do uso de analisador sintático em um agente conversacional. / Language is a mark of humanity and conscience, with the conversation (or dialogue) as one of the most fundamental manners of communication that we learn as children. Therefore one way to make a computer more attractive for interaction with users is through the use of natural language. Among the systems with some degree of language capabilities developed, the Eliza chatterbot is probably the first with a focus on dialogue. In order to make the interaction more interesting and useful to the user there are other approaches besides chatterbots, like conversational agents. These agents generally have, to some degree, properties like: a body (with cognitive states, including beliefs, desires and intentions or objectives); an interactive incorporation in the real or virtual world (including perception of events, communication, ability to manipulate the world and communicate with others); and behavior similar to a human (including affective abilities). This type of agents has been called by several terms, including animated agents or embedded conversational agents (ECA). A dialogue system has six basic components. (1) The speech recognition component is responsible for translating the user’s speech into text. (2) The Natural Language Understanding component produces a semantic representation suitable for dialogues, usually using grammars and ontologies. (3) The Task Manager chooses the concepts to be expressed to the user. (4) The Natural Language Generation component defines how to express these concepts in words. (5) The dialog manager controls the structure of the dialogue. (6) The synthesizer is responsible for translating the agents answer into speech. However, there is no consensus about the necessary resources for developing conversational agents and the difficulties involved (especially in resource-poor languages). This work focuses on the influence of natural language components (dialogue understander and manager) and analyses, in particular the use of parsing systems as part of developing conversational agents with more flexible language capabilities. This work analyses what kind of parsing resources contributes to conversational agents and discusses how to develop them targeting Portuguese, which is a resource-poor language. To do so we analyze approaches to the understanding of natural language, and identify parsing approaches that offer good performance, based on which we develop a prototype to evaluate the impact of using a parser in a conversational agent.

linguagem é uma marca da humanidade e da consciência, sendo a conversação (ou diálogo) uma das maneiras de comunicacão mais fundamentais que aprendemos quando crianças. Por isso uma forma de fazer um computador mais atrativo para interação com usuários é usando linguagem natural. Dos sistemas com algum grau de capacidade de linguagem desenvolvidos, o chatterbot Eliza é, provavelmente, o primeiro sistema com foco em diálogo. Com o objetivo de tornar a interação mais interessante e útil para o usuário há outras aplicações alem de chatterbots, como agentes conversacionais. Estes agentes geralmente possuem, em algum grau, propriedades como: corpo (com estados cognitivos, incluindo crenças, desejos e intenções ou objetivos); incorporação interativa no mundo real ou virtual (incluindo percepções de eventos, comunicação, habilidade de manipular o mundo e comunicar com outros agentes); e comportamento similar ao humano (incluindo habilidades afetivas). Este tipo de agente tem sido chamado de diversos nomes como agentes animados ou agentes conversacionais incorporados. Um sistema de diálogo possui seis componentes básicos. (1) O componente de reconhecimento de fala que é responsável por traduzir a fala do usuário em texto. (2) O componente de entendimento de linguagem natural que produz uma representação semântica adequada para diálogos, normalmente utilizando gramáticas e ontologias. (3) O gerenciador de tarefa que escolhe os conceitos a serem expressos ao usuário. (4) O componente de geração de linguagem natural que define como expressar estes conceitos em palavras. (5) O gerenciador de diálogo controla a estrutura do diálogo. (6) O sintetizador de voz é responsável por traduzir a resposta do agente em fala. No entanto, não há consenso sobre os recursos necessários para desenvolver agentes conversacionais e a dificuldade envolvida nisso (especialmente em línguas com poucos recursos disponíveis). Este trabalho foca na influência dos componentes de linguagem natural (entendimento e gerência de diálogo) e analisa em especial o uso de sistemas de análise sintática (parser) como parte do desenvolvimento de agentes conversacionais com habilidades de linguagem mais flexível. Este trabalho analisa quais os recursos do analisador sintático contribuem para agentes conversacionais e aborda como os desenvolver, tendo como língua alvo o português (uma língua com poucos recursos disponíveis). Para isto, analisamos as abordagens de entendimento de linguagem natural e identificamos as abordagens de análise sintática que oferecem um bom desempenho. Baseados nesta análise, desenvolvemos um protótipo para avaliar o impacto do uso de analisador sintático em um agente conversacional. / Language is a mark of humanity and conscience, with the conversation (or dialogue) as one of the most fundamental manners of communication that we learn as children. Therefore one way to make a computer more attractive for interaction with users is through the use of natural language. Among the systems with some degree of language capabilities developed, the Eliza chatterbot is probably the first with a focus on dialogue. In order to make the interaction more interesting and useful to the user there are other approaches besides chatterbots, like conversational agents. These agents generally have, to some degree, properties like: a body (with cognitive states, including beliefs, desires and intentions or objectives); an interactive incorporation in the real or virtual world (including perception of events, communication, ability to manipulate the world and communicate with others); and behavior similar to a human (including affective abilities). This type of agents has been called by several terms, including animated agents or embedded conversational agents (ECA). A dialogue system has six basic components. (1) The speech recognition component is responsible for translating the user’s speech into text. (2) The Natural Language Understanding component produces a semantic representation suitable for dialogues, usually using grammars and ontologies. (3) The Task Manager chooses the concepts to be expressed to the user. (4) The Natural Language Generation component defines how to express these concepts in words. (5) The dialog manager controls the structure of the dialogue. (6) The synthesizer is responsible for translating the agents answer into speech. However, there is no consensus about the necessary resources for developing conversational agents and the difficulties involved (especially in resource-poor languages). This work focuses on the influence of natural language components (dialogue understander and manager) and analyses, in particular the use of parsing systems as part of developing conversational agents with more flexible language capabilities. This work analyses what kind of parsing resources contributes to conversational agents and discusses how to develop them targeting Portuguese, which is a resource-poor language. To do so we analyze approaches to the understanding of natural language, and identify parsing approaches that offer good performance, based on which we develop a prototype to evaluate the impact of using a parser in a conversational agent.

As Zonas de Colaboração observadas a partir do movimento MetaReciclagem, uma rede que nasceu no campo da comunicação já no âmbito de uma cultura digital brasileira, em 2002, a partir da conversação na Internet de um grupo de pessoas sobre o desenvolvimento de blogs, reaproveitamento montagem/desmontagem de computadores, na interrelação entre essas ferramentas e a constituição de projetos colaborativos. O objeto de pesquisa é a relação comunicativa entre os membros do movimento que se contitui no ambiente virtual do MetaReciclagem. O método de pesquisa é a pesquisa-ação por meio de intervenção na lista de discussão do MetaReciclagem propondo uma metareflexão sobre as ações e resultantes do movimento ao longo dos anos. Abordagens quantitativa por meio de coleta de dados no ambiente virtual do MetaReciclagem (http://rede.metareciclagem.org) e qualitativa, coleta de depoimentos e conversações on-line dos membros que esboçaram suas opiniões e trocaram idéias a respeito do problema de pesquisa apresentado. Análise de dados por meio da técnica denominada análise de conteúdo e, dentro dessa, trabalhando com a técnica análise temática categorial. Os resultados demonstram que as Zonas de Colaboração são capazes de promover agenciamentos coletivos e intervenções sociais em políticas públicas de inclusão digital por meio das características da rede e do engajamento das pessoas. Tais transformações só acontecem pelo caráter não-hierárquico e não-institucionalizado do movimento do MetaReciclagem que promove a disseminação e replicação por meio de uma metodologia que privilegia a produção de subjetividade e a apropriação da tecnologia social em diferentes contextos culturais. / The Collaborative Zones observed in the MetaReciclagem (MetaRecycling Network) that arose out of the field of on-line communication, which was already enveloping Brazilian digital culture, in 2002, from a conversation on the internet among a group of people debating the development of blogs, reuse/recycling of computer parts, the interdependence of these tools and the constitution of collaborative projects. The purpose of this study is about the communication between the members of MetaRecycling Network (MetaReciclagem) and it\'s process as constituted in Collaborative Zones, result in actions of intervention in social situations in Brazil. This study proposes a research-action methodology, through intervention in the discussion lists of MetaRecycling, thus proposing a meta-reflection on the actions and results of the movement over the years. It involves quantitative (through the collection of data in the MetaRecycling virtual environment (http://rede.metareciclagem.org) and qualitative approaches, collecting on-line statements and conversations between members that sketch out their opinions and exchanging ideas regarding the issues of the presented research. It also involves data analysis by means of a technique entitled content analysis and, within this, working with categorical thematic technical analysis. The results have shown that Collaborative Zones are capable of leading to collective brokering and social interventions, as well as public policies involving digital inclusion through the characteristics of the network and consequent engagement with people. Such transformations can only happen because of the non-hierarchical and noninstitutionalized nature of the MetaRecycling movement, which promotes the dissemination and replication of methodology that privileges the production of subjectivity and the appropriation of social technology in different cultural contexts.

Работа посвящена изучению категориальных признаков дескриптивов винодельческой продукции в русской и англоязычных лингвокультурах на примере описаний органолептических характеристик вин. Материал исследования составили 250 текстов на английском языке и 250 текстов на русском языке, полученных на сайтах виноделен Великобритании, США, Канады, Австралии, Новой Зеландии и России. В первой части работы рассматриваются одни из ключевых понятий когнитивной лингвистики — концепт и категория. Изучаются характеристики дескриптива как средства категоризации и концептуализации, вводятся рабочие понятия категории, концепта и дескриптива, устанавливается взаимоотношение понятий категории и концепта, рассматривается роль вина в истории человечества и подходы к описанию вкусовых характеристик вин. Во второй части работы выявляются наиболее крупные концепты, к которым авторы описаний обращаются при описании запаха, вкуса, визуальных и тактильных характеристик вин, изучается работа этих концептов в их контекстуальном окружении. С помощью концептуального анализа были выявлены характеристики упоминаемых продуктов, к которым обращаются авторы описаний для передачи вкуса вина — так потенциальный покупатель может составить наиболее точный образ вкуса напитка в своем когнитивном пространстве. Наиболее частыми категориальными признаками, к которым происходит обращение, являются цвет продукта, особенности фактуры, род продукта, место произрастания, доступность продукта и др. В работе произведено описание выявленных категориальных признаков. На этапе сравнения признаков, обнаруженных в англоязычной и русскоязычной лингвокультурах, результаты исследования представляются в виде диаграмм. Результаты исследования могут быть использованы в маркетинговых целях и быть изучены с опорой на знания психолингвистики и лингвокультурологии. В приложении приводится список проанализированных текстов и сводные таблицы с результатами компонентного анализа дескрипций. / The work is devoted to the study of categorial features of descriptions of wine in Russian and English linguistic cultures based on the example of descriptions of organoleptic characteristics of wines. The research material consists of 250 texts in English and 250 texts in Russian, obtained from the websites of wineries in Great Britain, the USA, Canada, Australia, New Zealand and Russia. The first part of the work deals with the key concepts of cognitive linguistics – concept and category. The characteristics of description as a means of categorization and conceptualization are studied, the operational definitions of category, concept and descriptive are introduced, the relationship between the concepts of category and concept is established, the role of wine in human history and approaches to the description of wine taste characteristics are considered. The second part of the paper identifies the largest concepts to which the authors of the descriptions refer when describing the smell, taste, visual and tactile characteristics of wine, and studies the work of these concepts in their contextual environment. Conceptual analysis was used to identify the characteristics of the mentioned products that the authors of the descriptions refer to in order to convey the taste of wine so that the potential buyer can compose the most accurate image of the taste of the drink in their cognitive space. The most frequent categorial features that are referred to are product color, texture features, the genus of the product, the place of cultivation, the availability of the product, etc. The paper describes the identified categorical features. At the stage of comparing the categorial features found in the English and Russian-speaking linguocultures, the results of the study are presented in the form of diagrams. The results of the study can be used for marketing purposes and can be studied with the support of knowledge of psycholinguistics and linguoculturology. The appendix contains a list of the texts analyzed and summary tables with the results of the component analysis of the descriptions.

No description available.

No description available.

Combinatory categorial grammar (CCG) is a mildly-context sensitive formalism that is well-established in computational linguistics. At the basis of the grammar are a lexicon and a rule system: The lexicon assigns syntactic categories to the symbols of a given input string, and the rule system specifies how adjacent categories can be combined, yielding a derivation tree whose nodes are labeled by categories. In this thesis, we focus on composition rules, which are present in all variants of the grammar. Vijay-Shanker and Weir famously show that CCG can generate the same class of string languages as tree-adjoining grammar, linear indexed grammar, and head grammar. Their equivalence proof relies on two particular features of the grammar: ε-entries, which are lexicon entries for the empty word, and rule restrictions, which allow to restrict the rule set on a per-grammar basis. However, modern variants of CCG tend to avoid these features. This raises the question how this changes the generative and computational power of CCG. Another important feature is the rule degree, which determines how complex a certain category involved in a rule application may be. The goal of this thesis is to shed light on the effects that changing these features has. When modeling natural language, one is not only interested in the acceptability of a sentence, but also in its underlying structure. Therefore, we study the sets of constituency trees that CCG can generate, which are obtained by relabeling sets of derivation trees. We first provide a new proof of an analogous result by Buszkowski, showing that when only application rules are allowed, a proper subset of regular tree languages can be generated by CCG. Then, we show that when composition of first degree is included, CCG can generate exactly the regular tree languages. On the other hand, pure CCG, which allows all rules up to some degree, is shown to not even generate all local tree languages. Our main result on the generative capacity of CCG is its strong equivalence to tree-adjoining grammar. This means that these formalisms can generate the same class of tree languages. This is even the case when only composition rules of second degree and no ε-entries are used, showing that a CCG with these properties already has its full expressive power. Our constructions also provide an effective procedure for the removal of ε-entries. Regarding computational complexity, ε-entries and high rule degrees are in fact problematic. Kuhlmann, Satta, and Jonsson studied the universal recognition problem for CCG, which asks whether some given string is generated by some given grammar, considering both as part of the input. They prove that this problem is EXPTIME-complete if ε-entries are included, and NP-complete if not. We refine this result and show that the runtime is exponential only in the maximum rule degree of the grammar. Hence, when the rule degree is bounded by a constant, parsing becomes polynomial in the grammar size. This also holds when substitution rules are included in the rule system.

No description available.

The submitted doctoral thesis is an attempt to describe the nature and of the development of Brentano's theory of judgment. This description is followed by an introduction to the further development of Brentano's theory in the work of Brentano's most distinguished students Kazimierz Twardowski (1866-1938), Alexius Meinong (1853-1920), Anton Marty (1843-1914) and Edmund Husserl (1859-1938). The thesis is divided into five parts: The first part is dedicated to the explanation of Brentano's theory of judgment and starts with an interpretation of Brentano's two early books on Aristotle - On the several senses of Being in Aristotle (1862) and The Psychology of Aristotle (1867). The thesis presents Brentano's understanding of "being" in the sense of truth, his interpretation of the Aristotelian categories, his theory of parts and wholes, and his theory of intentionality and self-consciousness. Our interpretation then proceeds to Brentano's most known work, i.e. to Psychology from an Empirical Standpoint (1874), and presents the basic concept of this book, the concept of psychical phenomena. The resulting classification of psychical phenomena into three kinds introduces judgments as a kind of psychical phenomenon whose main feature consists in existential affirmation or rejection of an intentional object....