Apesar dos processos de recuperação e filtragem de informação sempre terem usado técnicas básicas de Processamento de Linguagem Natural (PLN) no suporte à estruturação de documentos, ainda são poucas as indicações sobre os avanços relacionados à utilização de técnicas mais sofisticadas de PLN que justifiquem o custo de sua utilização nestes processos, em comparação com as abordagens tradicionais. Este trabalho investiga algumas evidências que fundamentam a hipótese de que a aplicação de métodos que utilizam conhecimento linguístico é viável, demarcando importantes contribuições para o aumento de sua eficiência em adição aos métodos estatásticos tradicionais. É proposto um modelo de representação de texto fundamentado em sintagmas nominais, cuja representatividade de seus descritores é calculada utilizando-se o conceito de evidência, apoiado em métodos estatísticos. Filtros induzidos a partir desse modelo são utilizados para classificar os documentos recuperados analisando-se a relevância implícita no perfil do usuário. O aumento da precisão (e, portanto, da eficácia) em sistemas de Recuperação de Informação, conseqüência da pós-filtragem seletiva de informações, demonstra uma clara evidência de como o uso de técnicas de PLN pode auxiliar a categorização de textos, abrindo reais possibilidades para o aprimoramento do modelo apresentado / Although Information Retrieval and Filtering tasks have always used basic Natural Language Processing (NLP) techniques for supporting document structuring, there is still space for more sophisticated NLP techniques which justify their cost when compared to the traditional approaches. This research aims to investigate some evidences that justify the hypothesis on which the use of linguistic-based methods is feasible and can bring on relevant contributions to this area. In this work noun phrases of a text are used as descriptors whose evidence is calculated by statistical methods. Filters are then induced to classify the retrieved documents by measuring their implicit relevance presupposed by an user profile. The increase of precision (efficacy) in IR systems as a consequence of the use of NLP techniques for text classification in the filtering task is an evidence of how this approach can be further explored

Η εργασία αυτή κινείται στα όρια της μορφολογίας με τη σύνταξη και έχει ως απώτερο στόχο να προσδιορίσει τη θέση της μορφολογίας στο πλαίσιο της γραμματικής. Οι δομές που μελετώνται, τα χαλαρά πολυλεκτικά σύνθετα (βλ. Ράλλη, 2005, προσεχώς), έχουν τη μορφή [Ε Ο] και [Ο Ο σε γενική]. Ανήκουν στην κατηγορία των συνθέτων και ορίζονται ως μορφολογικοί σχηματισμοί. Σε αντίθεση όμως με τα κλασικά σύνθετα, εμφανίζουν συντακτικά χαρακτηριστικά. Στο πλαίσιο επομένως ενός δομικού συνεχούς (Ράλλη 2005) οι σχηματισμοί αυτοί θα μπορούσαν να τοποθετηθούν εντός της μορφολογίας, στα όρια όμως με τη σύνταξη. Στην εργασία αυτή θα αναλυθούν οι ξεχωριστές ιδιότητες των χαλαρών πολυλεκτικών συνθέτων και θα διαφοροποιηθούν τόσο από τις ονοματικές φράσεις, όσο και από τις ενδιάμεσες δομές. Η διαφοροποίηση των τριών ειδών δεν αποκλείει όμως την κοινή αναπαράστασή τους, με τη μορφή σχεδίων παραγωγής λέξεων, σε ένα ιεραρχημένο δομησιολόγιο, έτσι όπως προτείνεται από το θεωρητικό πλαίσιο της Δομησιακής Γραμματικής (Construction Grammar, Booij 2005). Στο ίδιο δομησιολόγιο εντάσσονται και οι σχηματισμοί της μορφής [Ο Ο σε ονομαστική], για τους οποίους υποστηρίζεται ότι βρίσκονται σε διαδικασία γλωσσικής αλλαγής και συγκεκριμένα μορφολογοποίησης. / The main object of my assignment is the analysis of loose multi-word compounds in Greek. There are two possible structures of multi-word compounds in Greek. So they appear combinations of adjective and noun, for example «μαύρη λίστα» (black list) and combinations of a noun in nominative and an other one in genitive, for example «μηχανικός αυτοκινήτων» (engineer of cars). It is argued that multi-word compounds are object of morphological analysis because they share a lot of common characteristics with the classical one-word compounds. I could mention their non compositional meaning. At the same time they share some syntactic features or properties. These compounds consist of two words like a syntactic construction. There is also agreement or inflection. So the object of this work is considered to be the interaction between morphology and syntax, because multi-word compounds in Greek are morphological constructions, in which syntax has some kind of authority. In this analysis are used some tests in order to distinguish similar formations, like noun phrases or compound-like phrases. Even though these kinds of structures do not belong in morphology, are represented in the same way in a hierarchical construction. The same fact occurs also for structures, which consist of two nouns both in nominative, for example «άνθρωπος αράχνη» (man spider). This kind of structure is considered to be under morphologization.

In this thesis the structure of the nominal group in Black South African English (BSAfE) is investigated by means of a comparison of data from the Tswana Learner English Corpus (TLEC) and the Louvain Corpus of Native English Essays (LOCNESS). Both corpora consist of student essays and are sub-corpora of the International Corpus of Learner English (ICLE). The TLEC represents a non-native variety of English, namely BSAfE, while LOCNESS represents native English from the United States and the United Kingdom. In the existing literature there are observations about and examples of (non-standard) characterizing features of BSAfE pertaining to nouns, determiners and quantifiers (e.g. Gough 1996), but until now, no in-depth study of the grammar of the nominal group in BSAfE has been undertaken. This study is an attempt to fill that gap. I present a description of the grammatical features of BSAfE observed in the corpus data in terms of linguistic functions and without assuming that they are errors or evidence of deficiencies. Though the approach is comparative (in the sense that a control corpus is used), it is primarily descriptive and non-normative, and as such, function-based. This study is conducted within the theoretical framework of functional linguistics, drawing on systemic functional linguistics as well as other functional and cognitive approaches to language. The specifying functions that the nominal group and its elements may fulfill form the basis of the descriptive framework. These functions are (1) the type-specifying function of the noun, (2) the (referent-)specifying function of the determiner, (3) quantification and (4) quality specification by modifiers and complements. The type-specifying function of the noun is particularly relevant in Chapter 4, which deals with the grammatical patterns of the noun people, although it also informs analyses in Chapters 5 and 6 which deal with determiners and quantification respectively. Quality specification is especially relevant in the description of the grammatical patterns of the noun people. The noun people is the most frequent noun in each of the corpora, but is at the same time a positive keyword in the TLEC, which means that it occurs much more frequently in the TLEC than in LOCNESS. Analysis of the full corpus concordances of this noun provides much evidence of anti-deletion in BSAfE (as first postulated by Mesthrie, 2006) and also sheds some light on left dislocation patterns involving the noun people (cf. Mesthrie, 1997). This analysis also reveals unique uses of the definite article and certain quantifying constructions in the TLEC data, which are investigated in the next two analysis chapters. A comparison of concordance samples of the articles indicates that the definite/indefinite distinction is made in both corpora and that there is not enough corpus evidence to postulate that there is a different system underlying the choice of article in BSAfE, such as a system based principally on the specific/non-specific distinction, as postulated by Platt, Weber and Ho (1984) for New Englishes in general. Analysis of the concordances of demonstrative and possessive determiners indicate that these determiners are used proportionally more frequently in the TLEC than in LOCNESS. Concordances of the words that are normally classified as quantifiers indicate that there are many more partitive-of quantifying constructions in the TLEC than in LOCNESS. The words some and most are positive keywords in the TLEC. After analyses of their concordances, it is concluded that their relative frequency can be attributed to the fact that some is often used merely as an indefinite marker and that most is often used as a synonym for many. The study shows that BSAfE largely shares its general grammar of the nominal group with other (including native) varieties of English, but at a finer level of analysis, some characteristic constructions and uses are detected. The corpus data indicate that the unique constructions in the TLEC data are mostly functionally motivated. These constructions represent conventionalized innovations in the sense used by Van Rooy (2010), rather than mere language learning errors. / Thesis (PhD (Linguistics and Literary Theory))--North-West University, Potchefstroom Campus, 2012

In this thesis the structure of the nominal group in Black South African English (BSAfE) is investigated by means of a comparison of data from the Tswana Learner English Corpus (TLEC) and the Louvain Corpus of Native English Essays (LOCNESS). Both corpora consist of student essays and are sub-corpora of the International Corpus of Learner English (ICLE). The TLEC represents a non-native variety of English, namely BSAfE, while LOCNESS represents native English from the United States and the United Kingdom. In the existing literature there are observations about and examples of (non-standard) characterizing features of BSAfE pertaining to nouns, determiners and quantifiers (e.g. Gough 1996), but until now, no in-depth study of the grammar of the nominal group in BSAfE has been undertaken. This study is an attempt to fill that gap. I present a description of the grammatical features of BSAfE observed in the corpus data in terms of linguistic functions and without assuming that they are errors or evidence of deficiencies. Though the approach is comparative (in the sense that a control corpus is used), it is primarily descriptive and non-normative, and as such, function-based. This study is conducted within the theoretical framework of functional linguistics, drawing on systemic functional linguistics as well as other functional and cognitive approaches to language. The specifying functions that the nominal group and its elements may fulfill form the basis of the descriptive framework. These functions are (1) the type-specifying function of the noun, (2) the (referent-)specifying function of the determiner, (3) quantification and (4) quality specification by modifiers and complements. The type-specifying function of the noun is particularly relevant in Chapter 4, which deals with the grammatical patterns of the noun people, although it also informs analyses in Chapters 5 and 6 which deal with determiners and quantification respectively. Quality specification is especially relevant in the description of the grammatical patterns of the noun people. The noun people is the most frequent noun in each of the corpora, but is at the same time a positive keyword in the TLEC, which means that it occurs much more frequently in the TLEC than in LOCNESS. Analysis of the full corpus concordances of this noun provides much evidence of anti-deletion in BSAfE (as first postulated by Mesthrie, 2006) and also sheds some light on left dislocation patterns involving the noun people (cf. Mesthrie, 1997). This analysis also reveals unique uses of the definite article and certain quantifying constructions in the TLEC data, which are investigated in the next two analysis chapters. A comparison of concordance samples of the articles indicates that the definite/indefinite distinction is made in both corpora and that there is not enough corpus evidence to postulate that there is a different system underlying the choice of article in BSAfE, such as a system based principally on the specific/non-specific distinction, as postulated by Platt, Weber and Ho (1984) for New Englishes in general. Analysis of the concordances of demonstrative and possessive determiners indicate that these determiners are used proportionally more frequently in the TLEC than in LOCNESS. Concordances of the words that are normally classified as quantifiers indicate that there are many more partitive-of quantifying constructions in the TLEC than in LOCNESS. The words some and most are positive keywords in the TLEC. After analyses of their concordances, it is concluded that their relative frequency can be attributed to the fact that some is often used merely as an indefinite marker and that most is often used as a synonym for many. The study shows that BSAfE largely shares its general grammar of the nominal group with other (including native) varieties of English, but at a finer level of analysis, some characteristic constructions and uses are detected. The corpus data indicate that the unique constructions in the TLEC data are mostly functionally motivated. These constructions represent conventionalized innovations in the sense used by Van Rooy (2010), rather than mere language learning errors. / Thesis (PhD (Linguistics and Literary Theory))--North-West University, Potchefstroom Campus, 2012

En el trabajo la semántica y pragmática de formas contraídas y no-contraídas en Alemán serán discutidas. Formas contraídas son preposiciones con sufijos flexionales, obligatorias en contextos en las que el contenido descriptivo del sintagma nominal es cierto de un individual o evento (expresiones que excluyen alternativas, referentes deducibles, Situative Unika, infinitivos nominalizados, referentes no específicos). Muchos autores consideran que las formas tienen artículos definidos que fusionan con una preposición. En contrario, propongo un análisis en términos de incorporación semántica en que una preposición tiene rasgos de caso, número y género. Interpretaciones únicas provienen del número singular y del contexto. Interpretaciones no específicas provienen de la semántica y el hecho de que el argumento nominal tiene menos rango que el argumento eventivo. / The semantics and pragmatics of contracted and non-contracted forms found in German will be discussed. Contracted form are prepositions with inflectional endings, and obligatory in contexts in which the descriptive content of the noun fits only one individual or event ("alternative-excluding" expressions, inferable referents, Situative Unika, nominalized infinitives, non-specific referents). Most accounts assume that contracted forms have underlying definite articles which have amalgamated with a preposition. In contrast, I propose to analyse these forms as semantically incorporating prepositions, which are inflected for (singular) number, gender, and case, and combine with noun phrases. Uniqueness effects are derived from singular number and from contextual entailments. Non-specific readings can directly be accounted for since the semantics predicts narrow scope of the nominal argument with respect to the event argument.

Studies on the phenomenon of hyperbaton (or discontinuous noun phrases) in Latin have traditionally suffered from over-generalizing descriptions and categorizations drawing on examples from a variety of sources. This thesis attempts a different approach, closely examining the totality of instances in a single text, Cicero’s Pro Milone, in view of establishing a preliminary typology and to investigate the motivations behind it. It also consistently compares the discontinuous data with continuous examples, in order to rule out as many alternative motivations as possible. The results suggest that the discursive value of the components is crucial for whether hyperbaton occurs or not: in particular a combination of a pragmatic function (Topic or Focus) and Contrast, and connects this to some recent theoretical proposals.

Submitted by Haroudo Xavier Filho (haroudo.xavierfo@ufpe.br) on 2016-05-19T18:03:16Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação. Dos Sintagmas Nominais aos Descritores Documentais Estudo de Caso na Indexação de Te.pdf: 3525255 bytes, checksum: 7cb6e4bda3f02eb17e2009285414748d (MD5) / Made available in DSpace on 2016-05-19T18:03:16Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Dissertação. Dos Sintagmas Nominais aos Descritores Documentais Estudo de Caso na Indexação de Te.pdf: 3525255 bytes, checksum: 7cb6e4bda3f02eb17e2009285414748d (MD5) Previous issue date: 2015-11-20 / O uso dos Sintagmas Nominais como instrumentos de organização e recuperação da informação digital vem se mostrando uma alternativa promissora para sistemas de informação. Nesse contexto, a indexação automática por meio de Sintagmas Nominais se mostra como uma alternativa que minimiza alguns problemas encontrados na indexação baseada em palavras isoladas, uma vez que os Sintagmas Nominais se constituem em unidades sintáticas que possuem semântica/sentido específico. No entanto, é notório que nem todos os Sintagmas Nominais que se encontram em um documento digital são representativos do mesmo, o que demonstra por sua vez a necessidade de uma seleção dos Sintagmas Nominais que realmente possam funcionar como descritores documentais. Nesse contexto, o presente trabalho tem como objetivo geral investigar a seleção de sintagmas nominais com valor de descritor no contexto do processo de indexação automática por meio de sintagmas nominais de resumos de teses e dissertações em português da área jurídica. Pretende-se: 1. Investigar o processo de indexação automática por meio de sintagmas nominais; 2. Verificar quais são as características de um Sintagma Nominal como valor de descritor documental; 3. Identificar na literatura científica nacional metodologias para seleção de sintagmas nominais em textos em português, bem como os critérios de seleção de cada metodologia; 4. Planejar experimento, onde os Sintagmas Nominais extraídos são categorizados quanto ao atendimento ou não a critérios de seleção propostos na literatura e quanto ao valor como Descritores, quando semelhantes aos descritores documentais resultantes da indexação manual; 5. Avaliar os critérios de seleção na indexação automática por meio de Sintagmas Nominais para teses e dissertações da área jurídica. Para o alcance dos objetivos propostos, fez-se uso de uma pesquisa bibliográfica e de um experimento. A pesquisa bibliográfica permitiu a identificação de pesquisas voltadas para a indexação automática por meio de Sintagmas Nominais, principalmente no que se refere à seleção de Sintagmas que funcionem como descritores documentais. Com base nas leituras dessas pesquisas, puderam-se identificar vários critérios utilizados para a seleção de Sintagmas. O experimento constituiu-se na aplicação dos critérios identificados na literatura aos sintagmas nominais extraídos de um conjunto de resumos de teses e dissertações da área jurídica, com o intuito de mensurar a utilidade ou não desses critérios no que diz respeito à seleção de sintagmas nominais descritores. Assim, realizou-se a indexação manual dos documentos, a extração automática dos Sintagmas Nominais constituintes dos mesmos, a categorização dos sintagmas nominais como descritores com base na semelhança aos descritores documentais advindos da indexação manual e a aplicação dos critérios de seleção aos sintagmas nominais extraídos. Por meio do experimento, foram percebidos comportamentos distintos entre cada critério, onde a maioria foi considerada útil para a seleção de sintagmas nominais. / The use of noun phrases as tools for information organization and retrieval has proven to be a promising alternative in information systems. In this context, automatic indexing through the noun phrases shows a new alternative that minimizes some problems encountered in indexing based on single words, given that noun phrases are syntactic units with specific semantic/meaning. However, it is evident that not all the noun phrases in a digital document are representative of the content, which in turn demonstrates the need for a selection of noun phrases that may serve as documentary descriptors. In this context, this work aims to investigate the selection of noun phrases with descriptor value in the context of the automatic indexing process by noun phrases from abstracts of theses and dissertations in the area of law in Portuguese. The specific objectives are: 1. Investigate the automatic indexing process by noun phrases; 2. Identify what are the characteristics of a noun phrase with descriptor value; 3. Identify the methodologies in national scientific literature for selection of noun phrases in texts in Portuguese, as well as the selection criteria of each methodology; 4. Organization of an experiment where each extracted noun phrase is categorized as descriptor and if it meets or not the proposed selection criteria in the literature; 5. Evaluate the selection criteria in the automatic indexing by noun phrases for theses and dissertations in the legal field. To achieve the objectives, it was made use of a bibliographic research and experiment. The literature review allowed the identification of research on automatic indexing by noun phrases, especially criteria used to choose noun phrases that act as documentary descriptors. Based on the readings of this research, it was possible to identify several criteria used for the selection of noun phrases. The experiment consisted in applying the criteria identified in the literature to noun phrases extracted from a set of abstracts of theses and dissertations in the legal field, in order to measure the usefulness or not of these criteria for the selection of noun phrases descriptors. Thus it was made manual indexing of documents, automatic extraction of noun phrases thereof, the categorization of noun phrases as descriptors based on their resemblance to the descriptors of manual indexing and the application of criteria in the extracted noun phrases. Through the experiment, we notice a different behavior for each criterion, where most of the criteria was considered useful for noun phrase selection.

One of the main characteristics of German technical language is the nominal style, which includes complex pre-nominal and post-nominal extended modifiers. A commonly held view is that these are less common in Swedish due to language-specific restrictions and preferences. As such, they may pose a challenge to Swedish translators. This essay examines this particular problem and focuses on the translation of four different complex extended modifiers: adjectival, participial, genitival and prepositional.The aim of this study was to determine which syntactic structures are used when these modifiers are translated into Swedish and to identify shifts using the concept “grammatical metaphor”, thereby focusing on the degree of grammatical metaphoricity. For the purposes of this study, a chapter of the technical book Neues Bauen mit Stroh in Europa by Gruber, Gruber and Sentler was translated into Swedish and then analysed with the above-mentioned aims in mind.The study showed that out of the 117 noun phrases with extended complex modifiers in the source text 21 were transposed into a less explicit, direct structure and more metaphorical language. The metaphorization was in some cases a result of simplification/omission of less dense semantic material and/or the translation into compounds. In 46 cases, the extended modifiers showed the same degree of grammatical metaphoricity as the source language expression and thus were re-metaphorized. In the remaining cases, a verbal or more explicit structure was chosen in the translation. This especially proved to be the case with pre-nominal extended adjectival and participial modifiers.

Apesar dos processos de recuperação e filtragem de informação sempre terem usado técnicas básicas de Processamento de Linguagem Natural (PLN) no suporte à estruturação de documentos, ainda são poucas as indicações sobre os avanços relacionados à utilização de técnicas mais sofisticadas de PLN que justifiquem o custo de sua utilização nestes processos, em comparação com as abordagens tradicionais. Este trabalho investiga algumas evidências que fundamentam a hipótese de que a aplicação de métodos que utilizam conhecimento linguístico é viável, demarcando importantes contribuições para o aumento de sua eficiência em adição aos métodos estatásticos tradicionais. É proposto um modelo de representação de texto fundamentado em sintagmas nominais, cuja representatividade de seus descritores é calculada utilizando-se o conceito de evidência, apoiado em métodos estatísticos. Filtros induzidos a partir desse modelo são utilizados para classificar os documentos recuperados analisando-se a relevância implícita no perfil do usuário. O aumento da precisão (e, portanto, da eficácia) em sistemas de Recuperação de Informação, conseqüência da pós-filtragem seletiva de informações, demonstra uma clara evidência de como o uso de técnicas de PLN pode auxiliar a categorização de textos, abrindo reais possibilidades para o aprimoramento do modelo apresentado / Although Information Retrieval and Filtering tasks have always used basic Natural Language Processing (NLP) techniques for supporting document structuring, there is still space for more sophisticated NLP techniques which justify their cost when compared to the traditional approaches. This research aims to investigate some evidences that justify the hypothesis on which the use of linguistic-based methods is feasible and can bring on relevant contributions to this area. In this work noun phrases of a text are used as descriptors whose evidence is calculated by statistical methods. Filters are then induced to classify the retrieved documents by measuring their implicit relevance presupposed by an user profile. The increase of precision (efficacy) in IR systems as a consequence of the use of NLP techniques for text classification in the filtering task is an evidence of how this approach can be further explored

No description available.