The use of gradable adjectives and adverbs represents an important means of expressing comparison in English. The grammatical forms of comparatives and superlatives are used to express explicit orderings between objects with respect to the degree to which they possess some gradable property. While comparatives are commonly used to compare two entities (e.g., “The blue whale is larger than an African elephant”), superlatives such as “The blue whale is the largest mammal” are used to express a comparison between a target entity (here, the blue whale) and its comparison set (the set of mammals), with the target ranked higher or lower on a scale of comparison than members of the comparison set. Superlatives thus highlight the uniqueness of the target with respect to its comparison set. Although superlatives are frequently found in natural language, with the exception of recent work by (Bos and Nissim, 2006) and (Jindal and Liu, 2006b), they have not yet been investigated within a computational framework. And within the framework of theoretical linguistics, studies of superlatives have mainly focused on semantic properties that may only rarely occur in natural language (Szabolsci (1986), Heim (1999)). My PhD research aims to pave the way for a comprehensive computational treatment of superlatives. The initial question I am addressing is that of automatically extracting useful information about the target entity, its comparison set and their relationship from superlative constructions. One of the central claims of the thesis is that no unified computational treatment of superlatives is possible because of their great semantic complexity and the variety of syntactic structures in which they occur. I propose a classification of superlative surface forms, and initially focus on so-called “ISA superlatives”, which make explicit the IS-A relation that holds between target and comparison set. They are suitable for a computational approach because both their target and comparison set are usually explicitly realised in the text. I also aim to show that the findings of this thesis are of potential benefit for NLP applications such as Question Answering, Natural Language Generation, Ontology Learning, and Sentiment Analysis/Opinion Mining. In particular, I investigate the use of the “Superlative Relation Extractor“ implemented in this project in the area of Sentiment Analysis/Opinion Mining, and claim that a superlative analysis of the sort presented in this thesis, when applied to product evaluations and recommendations, can provide just the kind of information that Opinion Mining aims to identify.

The present study investigated the '"fake-good" behaviour using the Superlative Scale S in theSpanish adaptation of the MMPI-2. From the 3.035 participants, the "normal" group ( 1.723)and the "clinic" group (561) responded with sincerity lo the questionnaire. A group of 269"'normal" participants was asked to show a negative image of themselves (group "fake-bad"')and other 282 participants, to show a positive image (group "Fake-good"); finally a group of 200 participants answered to the questionnaire without following the information of the items (group "inconsistent answers"'). The results showed the sensibility of this to detect the individuals that falsify their answers trying to present their more positive image. There were difficulties to identify individuals who answered to the questionnaire without keeping in mind the information of the items. / Se investigó la conducta de fingir en la adaptación española del MMPI-2. Participaron 3.035 personas, de los cuales los grupos "normal" (1.723) y "clínico" (561) respondieron con sinceiidad al cuestionario. A un grupo de 269 personas "normales'" se les instruyó para que muestren intencionadamente una imagen negativa de sí mismos (grupo '"mala imagen") y a otros 282,para que mostraran una imagen positiva (grupo "buena imagen"); otros 200 participantes contestaron al cuestionario sin atender a la información contenida en los ítems (grupo "repuestasinconsistentes"). Los resultados mostraron la sensibilidad de la Escala para detectar a aquellosque falsean sus respuestas presentando una imagen más favorable. Se observaron dificultades para discriminar a los que contestan sin tener en cuenta a los ítems.

This master thesis deals with the synthetic absolute superlative in contemporary Portuguese and its Czech counterparts. The aim was to describe the phenomenon in detail, explain the usage of doublet forms and discover the most frequent Czech counterparts of the absolute superlative in Portuguese language. The research was based on a corpora analysis of a Portuguese corpus named Corpus do Português and on a contrastive analysis of the parallel corpus Intercorp. The results show that in the case of the usage of the doublets, the borders between spoken and formal language might fade away, one of the doublets (formal vs. informal) often dominates and the usage is also influenced, apart from the level of formality, by other factors, for example by the variety of Portuguese. The most frequent Czech counterpart of the absolute superlative is a positive adjective usually bearing a higher value of expressiveness than the Portuguese adjective. The Czech language expresses the value of the absolute superlative primarily through its lexicon. Other counterparts are various expressions using adverbs, prefixes, diminutive or augmentative suffixes etc. In conclusion, a reverse control was made, and we identified the most frequent equivalents of several chosen Czech counterparts.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-05-25T11:08:51Z No. of bitstreams: 1 igordeoliveiracosta.pdf: 584668 bytes, checksum: 38d7187252785bd541b9fc96ddfacb3b (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-05-25T13:29:58Z (GMT) No. of bitstreams: 1 igordeoliveiracosta.pdf: 584668 bytes, checksum: 38d7187252785bd541b9fc96ddfacb3b (MD5) / Made available in DSpace on 2017-05-25T13:29:58Z (GMT). No. of bitstreams: 1 igordeoliveiracosta.pdf: 584668 bytes, checksum: 38d7187252785bd541b9fc96ddfacb3b (MD5) Previous issue date: 2010-08-31 / FAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais / Este trabalho tem como objeto a Construção Superlativa de Expressão Corporal (Construção SEC) (“solteirona e toda virgem, ignorava machezas, quase morreu de vergonha numa tarde de conversas”; “Padre Dito quase estourou de rir”; “O Lúcio rolou de rir com a explicação, e como consequência acabou virando a vítima e a cobaia do seminário”), um elo da grande rede de construções do Português denominadas por Miranda (2008a) como Construções Superlativas. Tendo como enfoque teórico a Linguística Cognitiva (FAUCONNIER, 1997; FAUCONNIER E TURNER, 2002; FILLMORE, 1977, 1982; FILLMORE E ATKINS, 1992; JOHNSON, 1987; LAKOFF, 1987, 1993; LAKOFF E JOHNSON, 1980, 1999; MIRANDA, 2002, 2008a, 2008b; SALOMÃO, 2002, 2008; dentre outros) e, mais especificamente, o seu Modelo Baseado no Uso da Gramática das Construções (CROFT AND CRUSE, 2004; GOLDBERG, 1995; LAKOFF, 1987; PERINI, 2001, 2008; TOMASELLO, 2003[1999]), propõe-se a descrição dos padrões formais e do valor semânticopragmático emergentes nas ocorrências dessa construção. Dado o relevo do uso no modelo teórico-analítico adotado, opta-se metodologicamente pela Linguística de Corpus (ALUÍSIO E ALMEIDA, 2006; SARDINHA, 2000, 2004), devido à possibilidade de se trabalhar com uma grande massa de dados naturais e de se verificar o grau de convencionalização/gramaticalização e produtividade da construção em foco (GOLDBERG, 1995). O corpus utilizado para pesquisa é o Corpus do Português (DAVIES E FERREIRA, 2006), composto por quarenta e cinco milhões de palavras, distribuídas em textos que perpassam os séculos XIV-XX. Partindo do trabalho de Sampaio (2007) em que parte da Construção SEC foi analisada dentro dos limites do campo metafórico da “morte” (morrer de rir, morrer de medo), nossas análises ampliam tal estudo, investigando a produtividade dessa rede (frequência de types) e a natureza do desencontro ou discrepância semântico e sintático (o fenômeno do mismatch) que institui, sincronicamente, seu padrão. Tal padrão – [Xv de YN/V] – tem em X um verbo (chorar de, rolar de, morrer de, se acabar de, se arrebentar de...) que representa o impacto físico ou fisiológico desencadeado pelo excesso de Y, um SN (de medo, de tristeza) ou um SV (de rir, de estudar). Nossas análises apontam ainda para: (1) a convencionalização da Construção SEC, que instancia, nos corpora investigados, 19 diferentes types; (2) a articulação do esquema imagético da Imposição de Forças (JOHNSON, 1987) – que, por sua vez, impõe uma Dinâmica de Forças (TALMY, 2000) –, a noção de Causalidade (LAKOFF E JOHNSON, 2002[1980]; TOMASELLO, 2003[1999]) e as metáforas primárias “Causa É Força Física” e “Intensidade É Escala” (LAKOFF E JOHNSON, 1999) como bases conceptuais do padrão construcional discutido; (3) a reanálise semântica do elemento X (verbos que suscitam os frames de impacto físico ou fisiológico) como um Operador Escalar, alinhando-se com outras construções canônicas existentes no Português, no frame de Escala Superlativa; (4) o uso pragmático da construção como estratégia argumentativa pertinente a contextos discursivos em que o falante/escritor possui maior liberdade de expressão subjetiva; e (5) a centralidade de processos figurativos na instituição de padrões gramaticais. / This paper focuses at the Superlative Construction of Body Expression (SEC Construction) (“solteirona e toda virgem, ignorava machezas, quase morreu de vergonha numa tarde de conversas”; “Padre Dito quase estourou de rir”; “O Lúcio rolou de rir com a explicação, e como consequência acabou virando a vítima e a cobaia do seminário”), a major link in the network of constructions of Portuguese named by Miranda (2008a) as Superlative Constructions. Having as theoretical basis the Cognitive Linguistics (FAUCONNIER, 1997; FAUCONNIER AND TURNER, 2002; FILLMORE, 1977, 1982; FILLMORE AND ATKINS, 1992; JOHNSON, 1987; LAKOFF, 1987, 1993; LAKOFF AND JOHNSON, 1980, 1999; MIRANDA, 2002, 2008a, 2008b; SALOMÃO, 2002, 2008; among others) and, more specifically, its Usage-based Model of Construction Grammar (CROFT AND CRUSE, 2004; GOLDBERG, 1995; LAKOFF, 1987; PERINI, 2001, 2008; TOMASELLO, 2003[1999]), it aims at the description of formal standards and the semanticpragmatic value emerging from the occurrences of this construction. Given the amount of use in the theoretical and analytical model adopted, the Corpus Linguistics (ALUÍSIO E ALMEIDA, 2006; SARDINHA, 2000, 2004) was the methodological approach chosen due to the possibility of working with a large mass of natural data and verification of the degree of conventionalization/grammaticalization and productivity of the construction focused (GOLDBERG, 1995). The corpus used for this research is the Corpus do Português (DAVIES AND FERREIRA, 2006), composed of forty-five million words, allocated in texts that pervade the XIV-XX centuries. Based on the work of Sampaio (2007), in which part of the SEC Construction was analyzed within the limits of the metaphorical field of "death" (morrer de rir, morrer de medo), our analysis extend such study, investigating the productivity of this network (frequency of types) and the nature of the semantic and syntactic disagreement or discrepancy (the phenomenon of mismatch) that establishes, synchronously, its pattern. This pattern – [Xv de YN/V] – has a verb as the X (chorar de, rolar de, morrer de, se acabar de, se arrebentar de...) that stands for the physical or physiological impact triggered by the excess of Y, a NP (de medo, de tristeza) or a VP (de rir, de estudar). Yet, our analysis points to: (1) a conventionalization of the SEC Construction, which instantiates 19 different types in the corpora investigated, (2) the articulation of the image schema of the Compulsion (JOHNSON, 1987) – which, in turn, requires the Force Dynamics (TALMY, 2000) –, the notion of Causality (LAKOFF AND JOHNSON, 2002[1980]; TOMASELLO, 2003[1999]) and the primary metaphors Causes Are Physical Forces and Intensity Is Scale (LAKOFF E JOHNSON, 1999) as conceptual basis of the constructional standard discussed; (3) the semantic reanalysis of X (verbs that raise the frames of physical or physiological impact) as a Scalar Operator, aligning itself with other canonical structures present in Portuguese, in the frame of Superlative Scale; (4) the pragmatic use of the construction as an argumentative strategy relevant to discursive contexts in which the speaker/writer has more freedom of subjective expression; and (5) the centrality of figurative processes in the institution of grammatical patterns.

In this thesis we concentrate on superlatives and other expressions to the greatest degree. The aim of this thesis is to map these expressions in the Spanish language of the youth and the Czech language of the youth. The thesis is divided to several parts. The first part describes expressions to the greatest degree, elatives, superlatives or other expressions of intensification from several different perspectives and the diversity of different theories represented by czech and spanish linguists. Then we mention global influence of the Language and we stipulate hypothesis. And in the second part we analyze czech and spanish magazines for young people, finding superlatives, elatives or other terms expressing the greatest degree. The result of this thesis is a well-arranged and organized report about the frequency, mood and fashion of different expressions of intensification, elatives or superlatives. In the conclusion we graphically demonstrate the frequency of different expressions and we prove the theoretical hypotheses.