The past two decades have shown an unexpected effectiveness of Web-scale data in natural language processing. Even the simplest models, when paired with unprecedented amounts of unstructured and unlabeled Web data, have been shown to outperform sophisticated ones. It has been argued that the effectiveness of Web-scale data has undermined the necessity of sophisticated modeling or laborious data set curation. In this thesis, we argue for and illustrate an alternative view, that Web-scale data not only serves to improve the performance of simple models, but also can allow the use of qualitatively more sophisticated models that would not be deployable otherwise, leading to even further performance gains. / Engineering and Applied Sciences

The primary aim of this study is to determine the level of syntactic development in English of South African matriculants. The ESL standard in South Africa has been criticised, but no objective data are available. This study provides relevant data, based on an index, and indicates shortcomings in learners' syntactic competence. Poor school-leaving standards in English are a cause of great concern in South Africa, and complaints about school leavers' standard of English have increased over the last few years. Yet, apart from generally being labelled as "poor", little is known about the actual level of development reached by ESL learners. Comments are often based on subjective impressions. This study focuses on syntactic development in writing and aims to determine the level of syntactic development of Grade 12 ESL learners in an objective way. Interlanguage, the concept of 'stages of development’ and fossilization are discussed. The need for an index that can measure language development objectively is considered. General means of measuring syntactic development are evaluated and an index formula is established by means of statistical analyses. This formula is based on the T-unit and assigns numerical values to levels of development. The index formula is used to determine the level of syntactic development of a group of Grade 12 ESL learners. The compositions that were analysed were obtained from six provinces in South Africa. Index values are calculated for Higher and Standard Grade, for the group as a whole, and each of the six provinces. An Error Analysis is conducted and error frequencies are reported. Problem areas in syntax are identified. The implications of the findings are considered and recommendations are briefly made for the teaching and learning of grammar. / Thesis (M.A. (English))--North-West University, Potchefstroom Campus, 2005.

The present paper is an attempt to examine the functional potential of deverbal abstract nouns in the clause. The verbal to nominal transfer has long been recognized as an important phenomenon in English. However, the peculiarities of usage of deverbal abstract nouns as well as their potential for expressing a range of semantic and syntactic functions have been little analysed in linguistic literature. The paper examines the concept of deverbal abstract nouns as non-congruent representation of processes, provides an overview of semantic and syntactic functions in the clause, and presents a detailed inventory of the range of semantic functions expressed by deverbal abstract nouns and their syntactic realization in the clause based on the linguistic evidence drawn from the corpus under investigation. The research was based on the semantic approach for interpreting the clause as representation of a process suggested by systemic functional linguistics. The examination of the scientific discourse text determined that deverbal abstract nouns were used in clauses of all types of processes, the most frequent being material and relational types; the semantic potential of deverbal abstract nouns was not limited to specific semantic functions as they were used to perform a full range of semantic functions in the clause; the most frequent semantic roles expressed by deverbal abstract nouns were those of the Circumstance, the Carrier and the Affected; the syntactic potential of deverbal... [to full text]

Das Herbsttreffen Patholinguistik wird seit 2007 jährlich vom Verband für Patholinguistik e.V. (vpl) durchgeführt. Das 6. Herbsttreffen mit dem Schwerpunktthema "Labyrinth Grammatik: Therapie von syntaktischen Störungen bei Kindern und Erwachsenen" fand am 17.11.2012 in Potsdam statt. Im vorliegenden Tagungsband finden sich alle Beiträge der Veranstaltung: die vier Hauptvorträge zum Schwerpunkthema, die Vorträge aus Praxis und Forschung von vier Patholinguistinnen in der Reihe Spektrum Patholinguistik sowie die Abstracts der Posterpräsentation. / The 'Herbsttreffen Patholinguistik' is an annual conference organized by the Association for Patholinguistics (Verband für Patholinguistik e.V./vpl) since 2007. The 6th Herbsttreffen with its theme "Labyrinth Grammar: Therapy of Syntactic Disorders in Children and Adults" took place on November 17th, 2012 in Potsdam. These proceedings contain all contributions of the meeting: the four keynote talks, the talks on speech/language therapy and research in the section "Spektrum Patholinguistik" by four patholinguists as well as the abstracts of the poster session.

Šiame darbe analizuojamas sakinio sandaros ir turinio pamatą sudarantys dviejų skirtingo tipo kalbų sintaksiniai predikatai ir jų struktūrinių tipų morfosintaksinis realizavimas daug tyrinėtas anglų kalbos gramatikos darbuose, tačiau ligi šiol mažai aptartas lietuvių kalbos moksliniuose techniniuose tekstuose. Sintaksinio predikato pamatas negali būti vertinamas tik funkciniu požiūriu. Labai svarbus veiksmažodžio (SP) struktūrinės sandaros vertinimas, lemiantis sintaksinės struktūros specifiką, kai veiksmažodis (SP) skaidomas iš vidaus (pvz., pagrindas [head] ir priklausomi rodikliai, linksniai). Veiksmažodžio sandara taip pat turi ir morfologinių, ir sintaksinių aspektų. / The predicate constitutes the basis of a sentence structure and content. The predicate has been researched at length and in depth in the works of English grammars, but is still little discussed in Lithuanian academic works, especially in those concerning scientific and technical texts. The structure of the predicate cannot be approached only functionally. The characterization of the internal division and morphosyntactic structure of the verb as the predicate is quite challenging (e.g. the verb as the head and affixes as dependent indicators in the VP). Accordingly, the morphosyntactic expression of a verb form and the structure inside the predicate is very complicated.

The predicate constitutes the basis of a sentence structure and content. The predicate has been researched at length and in depth in the works of English grammars, but is still little discussed in Lithuanian academic works, especially in those concerning scientific and technical texts. The structure of the predicate cannot be approached only functionally. The characterization of the internal division and morphosyntactic structure of the verb as the predicate is quite challenging (e.g. the verb as the head and affixes as dependent indicators in the VP). Accordingly, the morphosyntactic expression of a verb form and the structure inside the predicate is very complicated. / Šiame darbe analizuojamas sakinio sandaros ir turinio pamatą sudarantys dviejų skirtingo tipo kalbų sintaksiniai predikatai ir jų struktūrinių tipų morfosintaksinis realizavimas daug tyrinėtas anglų kalbos gramatikos darbuose, tačiau ligi šiol mažai aptartas lietuvių kalbos moksliniuose techniniuose tekstuose. Sintaksinio predikato pamatas negali būti vertinamas tik funkciniu požiūriu. Labai svarbus veiksmažodžio (SP) struktūrinės sandaros vertinimas, lemiantis sintaksinės struktūros specifiką, kai veiksmažodis (SP) skaidomas iš vidaus (pvz., pagrindas [head] ir priklausomi rodikliai, linksniai). Veiksmažodžio sandara taip pat turi ir morfologinių, ir sintaksinių aspektų.

Magistro darbo objektas – sakytiniai vakarų aukštaičių šiauliškių patarmės tekstai. Darbe analizuojama trijų pasirinktų tarminių tekstų sintaktika (iš adresato perspektyvos). Darbo pradžioje aptariami sakytinio teksto ir tarminės kalbos bruožai. Pasakotojų tekstai formaliai ir pagal prasmę yra siejami įvairiomis kalbinėmis ir nekalbinėmis priemonėmis. Dėmesys kreipiamas į tiriamų tekstų ekspresyvumą, gyvumą lemiančius kalbos elementus, sintaksines konstrukcijas: analizuojama tekstą sudarančių sakinių struktūra, jų ilgis, nagrinėjamos tarminei kalbai būdingos sintaksinės figūros. Tekstų rišlumas kuriamas tam tikromis sakinių siejimo priemonėmis: aptariamas elipsės, žodžių inversijos, suderintų veiksmažodžio laikų (visa tai sudaro gramatinę koheziją), įvairių kartojimų, hiponimų ir hiperonimų (leksinė kohezija), jungtukų ir kitų jungiamąją funkciją atliekančių žodžių (dalelyčių, prieveiksmių ir kt.) vartojimas. Be minėtų kalbinių priemonių, kalbėtojoms kurti sklandų, vientisą, o kartu ir ekspresyvų, tekstą padeda ekstralingvistiniai veiksniai: aprašomi analizuojamuose tarminiuose tekstuose matomi intonacijos ir ritmo dalykai. Atkreipiamas dėmesys, kad dėl tarminės kalbos spontaniškumo, menko minties apgalvojimo kalboje gali rastis tam tikrų nesklandumų. Darbe aptariamas vienas ryškiausių analizuojamos kalbos nesklandumo atvejų – nereikšminiai kartojimai. / The object of the master work is the spoken dialectal texts. There are analyzed the syntactic of three dialectal texts in the thesis. At first, the author describes the features of spoken text and dialectal speech. Formally and meaningfully narrators' texts are related to different linguistic and nonlinguistic devices. The attention is paid to the elements of speech, syntactic structure: analyzed sentences structure and length, syntactic shapes which are characteristic of dialectal speech. Cohesion of dialectal texts is determined for different connecting devices: the ellipsis, inverted order of words, verb tense agreement (grammatical cohesion), various reduplications, hyponyms and hyperonyms (lexical cohesion), connecting words such as conjunctions, particles, adverbs (lexical grammatical cohesion). In addition, the fluent speech also is created with nonlinguistic devices. There is described falling and rising intonation, speech rhythm in the master work. Also the master's author pays attention to nonmeaningful reduplications which is defined as nonfluencies of dialectal speech.

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.

This study investigated the lifespan influences of task type and working memory on the syntactic complexity of narrative discourse production. Participants included 180 healthy adults across three age cohorts: 20-29 years (Young Group), 60-69 years (Older Group) and 75-89 years (Elderly Group). Participants completed standardized working memory measures and four discourse tasks (single/sequential picture description, storytelling and personal recount). Syntactic complexity for each sample was measured via clausal density yielding a complexity index. For analysis, participants were placed into one of two groups based on working memory scores above (High Working Memory Group) or below (Low Working Memory Group) the mean. Significant differences in syntactic complexity between working memory groups were found for the single picture description and the storytelling; individuals in the high working memory group produced language with greater syntactic complexity. When the effects of cohort and working memory were investigated with a two-way ANOVA, working memory group was no longer significantly related to syntactic complexity. However, there was a significant relationship between cohort and syntactic complexity for the single picture description and storytelling tasks. Analyses indicate that the relationships between syntactic complexity, age, and working memory are dependent on task type.

The primary aim of this study is to determine the level of syntactic development in English of South African matriculants. The ESL standard in South Africa has been criticised, but no objective data are available. This study provides relevant data, based on an index, and indicates shortcomings in learners' syntactic competence. Poor school-leaving standards in English are a cause of great concern in South Africa, and complaints about school leavers' standard of English have increased over the last few years. Yet, apart from generally being labelled as "poor", little is known about the actual level of development reached by ESL learners. Comments are often based on subjective impressions. This study focuses on syntactic development in writing and aims to determine the level of syntactic development of Grade 12 ESL learners in an objective way. Interlanguage, the concept of 'stages of development’ and fossilization are discussed. The need for an index that can measure language development objectively is considered. General means of measuring syntactic development are evaluated and an index formula is established by means of statistical analyses. This formula is based on the T-unit and assigns numerical values to levels of development. The index formula is used to determine the level of syntactic development of a group of Grade 12 ESL learners. The compositions that were analysed were obtained from six provinces in South Africa. Index values are calculated for Higher and Standard Grade, for the group as a whole, and each of the six provinces. An Error Analysis is conducted and error frequencies are reported. Problem areas in syntax are identified. The implications of the findings are considered and recommendations are briefly made for the teaching and learning of grammar. / Thesis (M.A. (English))--North-West University, Potchefstroom Campus, 2005.