Wide-coverage parsing is an area that attracts much attention in natural language processing research. This is due to the fact that it is the first step tomany other applications in natural language understanding, such as question answering. Supervised learning using human-labelled data is currently the best performing method. Therefore, there is great demand for annotated data. However, human annotation is very expensive and always, the amount of annotated data is much less than is needed to train well-performing parsers. This is the motivation behind making the best use of data available. Turkish presents a challenge both because syntactically annotated Turkish data is relatively small and Turkish is highly agglutinative, hence unusually sparse at the whole word level. METU-Sabancı Treebank is a dependency treebank of 5620 sentences with surface dependency relations and morphological analyses for words. We show that including even the crudest forms of morphological information extracted from the data boosts the performance of both generative and discriminative parsers, contrary to received opinion concerning English. We induce word-based and morpheme-based CCG grammars from Turkish dependency treebank. We use these grammars to train a state-of-the-art CCG parser that predicts long-distance dependencies in addition to the ones that other parsers are capable of predicting. We also use the correct CCG categories as simple features in a graph-based dependency parser and show that this improves the parsing results. We show that a morpheme-based CCG lexicon for Turkish is able to solve many problems such as conflicts of semantic scope, recovering long-range dependencies, and obtaining smoother statistics from the models. CCG handles linguistic phenomena i.e. local and long-range dependencies more naturally and effectively than other linguistic theories while potentially supporting semantic interpretation in parallel. Using morphological information and a morpheme-cluster based lexicon improve the performance both quantitatively and qualitatively for Turkish. We also provide an improved version of the treebank which will be released by kind permission of METU and Sabancı.

The thesis deals with the issue of employees' satisfaction with individual areas of their work. The theoretical part describes the reasons for opinion ascertaining and factors influencing employee satisfaction and motivation. There are also listed the methods of employee satisfaction surveys and the approaches to statistical analysis of data from questionnaire surveys. The practical part is focused on the social climate analysis in the specific manufacturing company in areas such as remuneration, communication, work organization, teamwork, career growth, learning and development and other. The data analysis is based on categorial data analysis including detection of dependence in between individual answers and classification of employees, mainly by sex, age, department and position. For each area, there is a recommendation for the company management.

A primeira noção de topos, a de topos de Grothendieck, surgiu há cerca de 50 anos a partir de uma generalização do conceito de feixe na geometria algébrica. Poucos anos mais tarde, uma axiomatização categorial de algumas das propriedades de um topos de Grothendieck deu origem a uma segunda noção de topos, a de topos elementar; e essa descrição permitiu estabelecer ligações entre essas categorias e teoria dos conjuntos e lógica. Neste trabalho, estudamos a teoria de topos com um foco especial na construção da lógica interna dos topoi, e exploramos sua relação com modelos Heyting-valorados. / The first definition of a topos, that of a Grothendieck topos, emerged roughly 50 years ago from a generalization of the notion of sheaves in algebraic geometry. Few years later, a categorical axiomatization of some properties of Grothendieck topoi gave rise to a second notion of topoi, that of an elementary topos; and this description made it possible to establish connections between these categories and set theory and logic. In this work, we study topos theory with a particular focus on the construction of the internal logic of topoi, and explore its relation to Heyting-valued models.

FundaÃÃo de Amparo à Pesquisa do Estado do Cearà / CoordenaÃÃo de AperfeÃoamento de Pessoal de NÃvel Superior / Este trabalho visa a caracterizar formal, semÃntica e pragmaticamente as construÃÃes com verbo-suporte botar; considerando que, dentro desse âguarda-chuvaâ que se denominou verbo-suporte, hà estruturas com comportamentos sintÃtico-semÃnticos distintos. A partir das caracterÃsticas sintÃtico-semÃnticas das construÃÃes com verbo-suporte, verificaram-se trÃs graus de fluidez categorial: construÃÃes com verbo-suporte que estÃo mais prÃximas das construÃÃes livres (grau 1), construÃÃes com verbo-suporte consideradas prototÃpicas (grau 2) e construÃÃes que estÃo mais prÃximas das expressÃes cristalizadas (grau 3). A pesquisa enfocou o uso das construÃÃes botar + SN/SP em PortuguÃs e define as propriedades morfossintÃticas e semÃnticas que botar assume ao se vincular à categoria de verbo-suporte. A investigaÃÃo criteriosa sobre as propriedades de seleÃÃo de botar e seu comportamento sintÃtico-semÃntico em construÃÃes botar + SN/SP forneceu ainda subsÃdios para se descreverem diferentes empregos de botar nesse tipo de estrutura e, assim, se delinear uma cadeia de gramaticalizaÃÃo desse verbo. A base teÃrica linguÃstica de exame à a teoria funcionalista da linguagem, a qual reformula o corte rÃgido entre os verbos plenos e verbos-suporte, tratando esta categoria em uma perspectiva escalar e nÃo discreta. Os corpora (Norpofor, Porcufort e o DUP) de anÃlise compreendem ocorrÃncias de Vsup nas modalidades formal e informal do portuguÃs do Brasil, sem que se fixe como objetivo do trabalho pesquisar especificamente diferenÃas entre essas modalidades, mas com a hipÃtese de que a complexidade das CVSup nÃo poderia representar-se da mesma forma nessas modalidades de lÃngua. Realizou-se, com esta pesquisa, uma sistematizaÃÃo semÃntico-sintÃtica de expressÃes com verbo-suporte botar que apresentam graus de fluidez categorial. Para tanto, recorreu-se a anÃlises mÃltiplas que envolvem a descriÃÃo semÃntico-sintÃtica das expressÃes e a verificaÃÃo de parÃmetros que influenciam a fluidez e a depreensÃo de seus nÃveis. Os resultados demonstram ainda que a produtividade de botar na norma popular à maior do que na norma culta. No portuguÃs culto de Fortaleza, constatamos uma frequÃncia menor do verbo botar, confirmando nossa hipÃtese de que o processo de gramaticalizaÃÃo à mais lento na modalidade culta, embora, mesmo em menor quantidade, jà haja indÃcios de gramaticalizaÃÃo. ApÃs uma anÃlise geral nos sÃculos XVIII, XIX e XX, constatamos que hà um aumento da frequÃncia do verbo botar ao longo dos sÃculos. Tal fato confirma que esse verbo està em processo de gramaticalizaÃÃo contÃnuo. A descriÃÃo de cada um desses nÃveis (com os parÃmetros definidos na anÃlise e os exemplos extraÃdos dos corpora) explicitou que o verbo botar, na categoria de verbos-suporte, pode fazer parte tanto de estruturas mais integradas quanto de estruturas menos integradas, conforme essas construÃÃes se aproximam ou se distanciam do protÃtipo de uma construÃÃo com verbo-suporte. / This work aims to provide some formal, semantic and pragmatic characterization of the support verb âbotarâ, since the wideness concerning this verb presents distinctive syntactical and semantical behaviors in its structure. From the characteristics of the syntactical and semantical constructions of the verb, three degrees of categorical flushness were explored: free constructions, prototypical constructions and structures which are nearly crystalize. The research was concentrated in the constructions âbotar+SN/SPâ in Brazilian Portuguese, establishing the morpho-syntactical and semantical which the verb âbotarâ assumes when it is constructed as a support verb. The rigorous investigation of the selection properties of âbotarâ and its syntactical and semantical behavior in the âbotar+SN/SPâ structure provided support to describe different employments of the verb, hence leading to a chain of grammatical structures. The linguistic theoretical base used is the functionalism theory of language, which allows to precisely separating ordinary verbs from the support verbs, by treating their aspects in a scalar, rather discrete perspective. The corpus used in the analysis consists of Vsup occurrences in formal and informal Brazilian Portuguese language, however without restricting to only the differences among these modalities, but also using the hypothesis that the CVSup complexity could not be represented in these language modalities. The research allowed to create a systematic syntactical and semantical of expression involving the support verb which present degrees of categorical flushness. In order to achieve that, multiple analyses were conducted in which the syntactical and semantical descriptions of the expressions, in addition to the use of parameters which influences the flushness and the comprehension. As a result of the research, the productivity of the verb is more evident in the popular rather than in the formal usage. In particular, in formal Portuguese of Fortaleza, we verified a lower frequency of use, which confirms our starting hypothesis that the grammaticalization is slower in the formal language, although we could note that it has started already. After detailed analyses in the XVIII, XIX and XX centuries, we concluded that the usage of the verb has been increasing in the past few centuries, in fact the verb is having a gradual continuous grammaticalization. According to the analysis parameters and the examples obtained from the corpus, the description of each of those levels showed that the verb âbotarâ, within the support verb category, can be part of both the less and more integrated structures, depending on the how close those constructions are from the prototype of a support verb construction.

This study investigates the problem of quantifier scope ambiguity in natural languages and the various ways with which it has been accounted for, some of which are problematic for monotonic theories of grammar like Combinatory Categorial Grammar (CCG) which strive for solutions that avoid non-monotonic functional application, and assume complete transparency between the syntax and the semantics interface of a language. Another purpose of this thesis is to explore these proposals on examples from Turkish and to try to account for the meaning differences that may be caused by word order and see how the observations from Turkish fit within the framework of CCG.

This thesis investigates fundamental questions surrounding the phenomenon of control, with an emphasis on control in Turkish, as well as the behaviour of control verbs in non-infinitival environments, which have received little attention previously. I focus solely on the cases of obligatory control (OC) which constitute the only kind of control that is conditioned by the matrix verb alone. This approach is couched in Combinatory Categorial Grammar (CCG) where the control verb projects the necessary syntactic and semantic information. In particular, I argue that the control behaviour is an entailment associated with the verb itself, and that variable, split and partial control are instances of OC. Hence, no special mechanism/structure is needed to account for their interpretation. As to the syntactic and semantic status of the complement, I maintain that the complement is a bare VP in syntax and denotes a property in semantics. Building upon the conclusions reached about OC, I attempt to account for additional complementation patterns of OC verbs. I argue that here too the matrix verb has a crucial role in ruling in and out possible complement types. Finally, I note that control involves much more than just figuring out the reference of the &ldquo / unexpressed&rdquo / subject of the complement, and I furthermore propose that the additional frames of an OC verb provide important clues as to its lexical meaning, which are argued to be relevant for the acquisition of control.

A large proportion of computationally-oriented theories of grammar operate within the confines of monostratality (i.e. there is only one level of syntactic analysis), compositionality (i.e. the meaning of an expression is determined by the meanings of its syntactic parts, plus their manner of combination), and adjacency (i.e. the only operation on terminal strings is concatenation). This thesis looks at two major approaches falling within these bounds: that based on phrase structure grammar (e.g. Gazdar), and that based on categorial grammar (e.g. Steedman). The theories are examined with reference to extraction and coordination constructions; crucially a range of 'compound' extraction and coordination phenomena are brought to bear. It is argued that the early phrase structure grammar metarules can characterise operations generating compound phenomena, but in so doing require a categorial-like category system. It is also argued that while categorial grammar contains an adequate category apparatus, Steedman's primitives such as composition do not extend to cover the full range of data. A theory is therefore presented integrating the approaches of Gazdar and Steedman. The central issue as regards processing is derivational equivalence: the grammars under consideration typically generate many semantically equivalent derivations of an expression. This problem is addressed by showing how to axiomatise derivational equivalence, and a parser is presented which employs the axiomatisation to avoid following equivalent paths.

The diploma thesis describes the selected event nominals in the legal texts specifically in the corpus named CORTE. The approaches of these nominals were selected in the relevant bibliography and the diploma thesis focuses especially on the themes and on the criterions which are used to define the event nominals. On the basis of these definitions the selected event nominals will be assess with regard to their action or non-action meanings, valency and colocability with verbs and also to their semantic properties. Furthermore, it is analyzed the relationship between the base and the derivation suffix on the one hand and semantic properties of the event nominals on the other. The corpus CORTE is used for the analysis, SketchEngine, web interface, and program Paraconc serve to find out the information about event nominals.

Cette thèse explore l'usage des Grammaires Categorielles Abstraites (CGA) pour la Génération Automatique de Texte (GAT) dans un contexte industriel. Les systèmes GAT basés sur des théories linguistiques ont un long historique, cependant ils sont relativement peu utilisés en industrie, qui préfère les approches plus "pragmatiques", le plus souvent pour des raisons de simplicité et de performance. Cette étude montre que les avancées récentes en linguistique computationnelle permettent de concilier le besoin de rigueur théorique avec le besoin de performance, en utilisant CGA pour construire les principaux modules d'un système GAT de qualité industrielle ayant des performances comparables aux méthodes habituellement utilisées en industrie. / This thesis explores the usage of Abstract Categorial Grammars (ACG) for Natural Language Generation (NLG) in an industrial context. While NLG system based on linguistic theories have a long history, they are not prominent in industry, which, for the sake of simplicity and efficiency, usually prefer more ``pragmatic" methods. This study shows that recent advances in computational linguistics allow to conciliate the requirements of soundness and efficiency, by using ACG to build the main elements of a production grade NLG framework (document planner and microplanner), with performance comparable to existing, less advanced methods used in industry

Research into parsing sign language corpora is ongoing. Corpora for German Sign Language and Italian Sign Language have been parsed (Bungeroth et al., 2006; Mazzei, 2011, 2012, respectively). However, research into parsing a corpus of American Sign Language is non-existent. Examples of parsed ASL sentences in literature are typically isolated examples used to show a particular type of construction. Apparently no attempt has been made to parse an entire corpus of American Sign Language utterances. This thesis presents a method for constructing a grammar so that a parser implementing Combinatory Categorial Grammar can parse a corpus of American Sign Language. The results are evaluated and presented.