• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 236
  • 74
  • 58
  • 29
  • 17
  • 15
  • 14
  • 11
  • 8
  • 8
  • 6
  • 4
  • 4
  • 4
  • 2
  • Tagged with
  • 551
  • 360
  • 115
  • 88
  • 81
  • 68
  • 67
  • 57
  • 54
  • 47
  • 46
  • 44
  • 42
  • 41
  • 41
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
41

Phonological mismatches: how does the position and degree of the mismatch affect spoken word recognition?

Tracy, Erik Charles 25 September 2006 (has links)
No description available.
42

Recognition of code-switched words by Spanish-English bilinguals

Castellon Perez, Yvonne 01 July 2001 (has links)
No description available.
43

Reticence, anxiety and performance of Chinese university students in oral English lessons and tests. / CUHK electronic theses & dissertations collection / Digital dissertation consortium

January 2005 (has links)
Adopting both qualitative and quantitative approaches with a strong element of triangulation, the present research aimed to investigate the issues of reticence and anxiety in oral English classroom-learning and testing situations at the tertiary level in Mainland China. / Following the ground work phase, the main study, which was divided into two phases, got underway. In the first phase, a 124-item survey was distributed to approximately 570 first-year undergraduate non-English majors at three different proficiency levels at Tsinghua University; 547 valid questionnaires were processed. In the second phase, three English language classes (one from each different proficiency group) were selected for the case study which continued for the full term: the students were required to write reflective journals with one entry per week for six consecutive weeks, the teachers were asked to observe and keep a weekly record of the students' reticence and anxiety in classroom activities during the whole term, the three classes were observed and video-taped twice, 7 students from each group and their teachers were invited for semi-structured interviews, and the final oral English test given to the students was observed and video-taped. / The large-scale survey was analyzed using SPSS 11.00---descriptive statistics, reliability, correlation, factor analysis, t-test, and ANOVA to examine students' self-reported reticence and anxiety and their relationships with other variables such as students' self-rated English proficiency. The journals, interviews, and observations were subjected to a thematic content analysis to explore students' reticence and anxiety during oral English lessons and tests, and identify factors underlying reticence and anxiety in oral English classroom-learning and testing situations as well as corresponding coping strategies adopted by the participants. / The major findings were: (1) a considerable number of students self-reported and were observed to be reticent and nervous during oral English lessons and tests. The more proficient students tended to be less reticent and anxious, (2) the more reticent student tended to be more anxious during oral English lessons and tests, (3) reticence and anxiety negatively affected students' performance in oral English, (4) student reticence and anxiety varied from activity to activity in the classroom and changed during both the term and final oral English test, (5) multiple factors contributed to student reticence and anxiety during oral English lessons and tests, (6) the majority of the participants felt helpless about student reticence and anxiety. Most teachers and students were not aware of how to effectively cope with them, and (7) reticence and anxiety interacted with each other in both situations. Based on these findings, certain pedagogical implications were discussed to reduce students' reticence and anxiety in oral English classroom-learning and testing situations, thus enhancing the teaching and learning of oral English in Mainland China. In addition to contributing to the overall literature of research on reticence and anxiety in Chinese and FL learning contexts, the present research revealed some areas for future research. / Liu Meihua. / "July 2005." / Adviser: Jane Jackson. / Source: Dissertation Abstracts International, Volume: 67-01, Section: A, page: 0167. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 336-360). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.
44

Izincomo phakathi kwesiduna namabhinqa esiXhoseni

Ralarala, Knowledge Monwabisi 03 1900 (has links)
Thesis (MA)--University of Stellenbosch, 2003. / ENGLISH ABSTRACT: This study aims to investigate the speech act of complimenting in Xhosa. The study constitutes of six chapters with respective to execute this aim. Chapter one provides special details regarding the aim of this study, the method that has been followed, and the organisational structure of the study. Chapter two focuses on both speech act and politeness theories. The central notion relates to (a) the acts of locution, (b) illocution and (c) perlocution. These elements of speech acts entail the notion that utterances that are produced by participants in a conversation comprises of (a) the actual sounds and words uttered, and those words and sounds (b) are intended towards the fulfilment of the force or intention behind them and (c) the effect of that force is intended to the hearer. Although there are other related elements, this notion is prominent in this chapter. Chapter three examines speech acts of complimenting in Xhosa along with their responses. This examination is informed by various ideas from respective researchers. For an effective and successful investigation of speech acts, a guideline which serves as a base follows a model of ethnography of communication. Almost all these researchers are putting emphasis on this view. This chapter takes this issue further by exploring the differences in compliments and responses with reference to gender, along with other qualities that have been mentioned. The elements of the responses the principles, their nature, structure and appearance in general conversations with specific reference to complimenting, are other key properties examined in this chapter. Chapter four focuses on the functions of the compliments. For example, almost all the researchers in the field are in agreement that compliments serve to revitalize, establish or create or encourage solidarity. Although there are other functions relating to this speech act, such as replacing other conversational formulas, e.g. greetings, softening criticism, the function of solidarity is perceived to be central. Another area which receives attention is the structural qualities of the compliment, along with syntactic and lexical features. This analysis explores the syntactic categories that relate to this work, together with the formulaic nature of this speech act. In addition, the status and gender of the participants in an interaction, as informed by situations or friends who like to be complimented, are all matters that receive scrutiny in this chapter. Chapter five constitute the central part of the study. The differences in compliments and responses receive attention with the aim of establishing possible difference between compliments given by males and females. Chapter six is the last chapter, it presents the findings and conclusion of the study. / AFRIKAANSE OPSOMMING: Hierdie studie het as doelstelling die ondersoek van die spraakhandeling van komplimentering in Xhosa. Die tesis beslaan ses hoofstukke wat gestruktureer is om uitvoering te gee aan hierdie doelstelling. Hoofstuk een gee spesifieke besonderhede betreffende die doelstelling van die studie, die metode wat gevolg word, en die organisasie van die studie. Hoofstuk twee fokus op spraakhandelingsteorie en hoflikheidsteorie. Die sentrale konsep hou verband met (a) die handelinge van lokusie, (b) ilokusie en (c) perlokusie. Hierdie elemente van spraakhandelinge behels die begrip dat die uitinge wat deur deelnemers in 'n gesprek geproduseer word bestaan uit (a) die werklike klanke en woorde wat geuiter word, en wat (b) die intensie het om uitdrukking te gee aan die fors daaragter, en (c) die effek is gerig op die hoorder. Hierdie oorweging is sentraal in die hoofstuk. Hoofstuk drie ondersoek die spraakhandeling van komplimentering in Xhosa tesame met die response daarop. Hierdie ondersoek word ingelig deur sieninge van verskillende navorsers. Vir die effektiewe ondersoek van spraakhandelinge, is 'n rigtinggewende basis vanuit die etnografie van kommunikasie nodig. Hierdie hoofstuk ondersoek die verskille in komplimente en die response daarop met betrekking tot gender en ander eienskappe. Die aspekte van die response, hulle beginsels, aard, en struktuur, en die gebruik daarvan in gesprekke wat komplimente bevat, word ondersoek. Hoofstuk vier fokus op die funksies van komplimente. Feitlik al die navorsers in die veld van komplimentering stem saam dat komplimente die doel het om solidariteit te skep of te versterk. Alhoewel daar ander funksies is van komplimentering, bv. die verplasing van gespreksformules, soos groet, kritiek, is die bevestiging van solidariteit sentraal. 'n Ander area wat aandag kry in die hoofstuk, is die strukturele kenmerke van komplimente, insluitende sintaktiese en leksikale kenmerke. Die analises behels 'n ondersoek na die sintaktiese kategorieë wat komplimente realiseer, tesame met die formule-agtige aard van komplimente as spraakhandelinge. Voorts word die status en gender van die gespreksgenote ondersoek. Hoofstuk vyf vorm die sentrale deel van die studie betreffende die bepaling van moontlike gender verskille in komplimente. Hoofstuk ses bied die samevatting en gevolgtrekkings van die studie. / ISISHWANKATHELO: Esi sifundo sikwiphulo lokuphanda ngezenzo zentetho zezincomo. Kuqulunqwe izahluko ezintandathu ezineenjongo ezahlukiyo ukufezekisa eli phulo. Isahluko sokuqala sinika umkhombandlela malunga nenjongo yesifundo, imethodi elandelweyo kananjalo nolwakhiwo zinikwa ingqwalasela. Isahluko sesibini sinika ingqwalasela kwithiyori yezenzo zentetho kananjalo neyentlonelo. Uluvo olungundoqo nolunika umdla kule ngxelo luqulathe izenzo ze (a) ilokhushini, (b) 1- ilokhushini ne (c) nephelokhushini. Ezi zenzo zixhibe ingcamango yokuba intetho eveliswa ngabo bathatha inxaxheba kunxibelelwano iqulathe (a) izandi zolwakhiwo Iwesigama, ukuze eso sigama (b) sijoliswe kwinjongo ethile ekuyimfuneko ukuba iphunyezwe ekugqibeleni (c) ifuthe laloo mpumelelo lingqale lowo kunxityelelwano naye okanye umphulaphuli. Nasekubeni ekhona namanye amasolotya axhagileyo, kodwa le ngcamango iphawuliweyo iphume izandla kwesi sahluko. Isahluko sesithathu siphonononga izenzo zeentetho zezincomo kananjalo neempendulo zazo. Olu phononongo lungqongwe zizimvo zeengcali eziphuma kwiinkalo ngeenkalo. Umkhombandlela osisikhokhelo kwiphulo lophando Iwezenzo zeentetho naziphi na, ukuze lube yimpumelelo kwaye lusulungeke kufuneka luiandeie imodeli okanye umgaqo weethinografi yonxibelelwano phantse bonke aba baphandi babeke umnwe kule ngcamango. Ngokunxusa oku, esi sahluko, xa sihlabela mgama, siphawula ngeyantlukwano yezincomo neempendulo zazo ngesini, nangezinye iimpawu ezikhankanyiweyo. Uqulunkqo Iwemigaqo yokuphendula, ubunjani, ubume nembonakalo yayo entethweni gabalala ngokunxulumene nezincomo ngamanye amasolotya abalulekileyo nanika umdla kwesi sahluko. Isahluko sesine sinika ingqwalaselo kwimisebenzi yesincomo. Umzekelo, phantse zonke ezi ngcali zikweli phulo ziyayingqina nento yokuba izincomo zivuselela , ziyila okanye zikhuthaza umanyano. Nangona ibonakala ikhona eminye imisebenzi ekhankanywayo engqonge esi senzo sentetho njengokuthabatha indawo yezinye zeefomuia zonxulumano, umz. imibulisi, ukuthomakalisa ugxeko-ncomo, olu luvo lomanyano lubonwa njengoluhamba phambili. Enye indima ephononongwayo ziimpawu zolwakhiwo Iwezincomo, kananjalo neempawu zesintaksi neleksikali. Olu hlalutyo lulo oludulisa izigaba zentetho ezibandakanyekayo kananjalo nobufomula besi senzo sentetho. Ukwaleka, iwonga nesini sabo bathabatha inxaxheba kolu nxibelelwano ngokulawulwa ziimeko okanye izihlobo ezikholisa ukunconywa, konke oku yimibandela efumana ukuphicothwa kwesi sahluko. Isahluko sesihlanu siqulathe umongo wesi sifundo. Uhlalutyo Iwezincomo neempendulo zazo lunikwa ingqwalasela ngamehlo abukhali; ngeenjongo zokufumanisa umahluko ongaba ukhona phakathi kwezincomo zamaduna namabhinqa. Isahluko sesithandathu nesokugqibela sinika iziphumo zophando nesiphelo sesi sifundo.
45

The speech act of complaining in isiXhosa

Manjiya, Monelwa Precious 12 1900 (has links)
Thesis (MA)--Stellenbosch University, 2001. / ENGLISH ABSTRACT: This study investigates the speech act of complaining in Xhosa. It is organized into six chapters. Chapter 1 states the aim of the study. The method, design and the organization of the study are also presented. In Chapter 2 the speech acts and politeness theories are presented as the basic framework of this study. It is clear that people do not only produce utterances, which contain grammatical structures and words when attempting to express themselves, but they always perform actions via those utterances. The actions people perform via their utterances are done in accordance within a certain area of linguistic pragmatics. Chapter 3 deals with the speech act of complaining as discussed by various theorists. Chapter 4 is concerned with complaints strategies. Here it is revealed that one must be able to voice one's annoyance or anger while at the same time it is important to avoid embarrassment by creating a situation in which it becomes impossible for him/her to face the aversiveness of taking the blame. Chapter 5 deals with complaint situations in Xhosa. In order to make sense of what is said during an interaction, various factors have been examined, which relate to social distance and closeness. The relative status of the participants is based on social values. Complaint situations are divided into three social groups: power relations, friendship and strangers. Power relations involve people of superior and low status. This group is divided into two subgroups: people with superior status to people with low status and vice versa; those speakers with a lower status in a Xhosa speaking context tend to mark social distance between themselves and higher status speakers. Among friends, there is no social distance because people tend to treat one another as equals. The same is true of strangers, although they don't know each other; they tend to treat one another with mutual respect. Most of the time complaints in Xhosa express disapproval or negative feelings towards the source of the complaint. Complaints sometimes can be impolite. They can lead to conflict as they are sometimes very threatening, accusing and cursing. That is why a number of strategies have been introduced to a complainant who wants to avoid direct confrontation with the complainee. The indirect accusation strategy is the number one tool, which is always used by the complainers to avoid conflict, unlike annoyance, direct accusation and explicit blame of the accused's action or of the accused as a person. People who use indirect accusations do not want to run the risk of losing face, unlike the direct accusation, which is face-threatening. Responding to a complaint is also an important factor as it promotes further interaction. Response serves as a way of displaying interest in what the speaker is saying. It has been noticed that response gives the speakers the opportunity to voice their feelings. Finally, Chapter 6 presents the conclusions of the investigation and the main findings of the study are summarized. / AFRIKAANSE OPSOMMING: Hierdie studie ondersoek die spraakhandeling van klagte in Xhosa. Dit is georganiseer in ses hoofstukke. Hoofstuk 1 bied die doelstelling van die studie. Die metode, ontwerp en organisasie van die studie word ook bespreek. In Hoofstuk2 word die spraakhandelinge en beleefdheidsteorieë bespreek as die raamwerk van die studie. Dit word duidelik gemaak dat mense nie slegs uitinge produseer wat grammatikale strukture en woorde bevat nie, maar hulle voer altyd aksies uit via die uitinge. Hierdie aksies wat mense uitvoer via hulle uitinge word gedoen in ooreenstemming in bepaalde area van linguistiese pragmatiek. Hoofstuk 3 behandel in die besonder die spraakhandeling van klagte soos ondersoek deur verskillende navorsers. Hoofstuk 4 behandel klagte-strategieë. Hier word aangedui dat 'n spreker in staat moet wees om hy/sy se ontevredenheid of woede te lug terwyl dit terselfdertyd belangrik is om verleentheid te vermy deur die skep van 'n situasie waar dit vir hom/haar onmoontlik maak om te staan te kom voor die neem van die blaam. Hoofstuk 5 ondersoek klagte-situasies in Xhosa. Ten einde sin te maak oor wat gesê word en 'n interaksie, is verskeie faktore ondersoek wat verband hou met sosiale opstand en nabyheid. Die realtiewe status van die deelnemers word gebaseer op sosiale waardes. Klagte-situasies kan verdeel word in terme van drie sosiale groepe: magsverhoudinge, vriendskap en vreemdelingskap. Magsverhoudinge hou verband met die verhoudinge tussen mense met hoë en lae status. Hierdie groep word verdeel in twee sub-groepe: mense met superieuse status en mense met lae status en omgekeerd. Persone met 'n laer status in 'n Xhosasprekende konteks is geneig om sosiale afstand te markeer tussen hulleself en hoër status sprekers. Daar is geen sosiale opstand tussen vriende nie, omdat sulke persone geneig is om mekaar as gelykes te behandel. Dieselfde is waar van vreemdelinge, want alhoewel hulle mekaar nie ken nie, is hulle geneig om mekaar met respek te behandel. Die meeste van die klagtes oor tyd in Xhosa gee druk misnoeë of negatiewe gevoelens uit met betrekking tot die bron van die klagte. Klagtes kan soms onbeleefd wees. Hulle kan lei tot konflik, aangesien hulle soms baie bedreigend aantygend en beledigend kan wees. Daarom is 'n aantal strategieë beskikbaar vir 'n klaer wat direkte konfrontasie wil vermy teenoor die persoon waarteen gekla word. Die indirekte aantyging strategie is die voorkeur strategie wat deur die klaers gebruik word kom konflik te vermy, in teenstelling met kwaad word direkte aantyging en eksplisiete blaam van die aangetygde persoon se aksie of die persoon self. Persone wat indirekte aantygings gebruik, wil nie die risiko loop om "gesig" (face) te verloor nie, in teenstelling met die direkte aantyging, wat gesig-bedreiging is. Om te reageer op 'n klagte, is ook 'n belangrike faktor, aangesien dit verdere interaksie aanmoedig. Die gee van 'n reaksie is 'n wyse om aan te toon dat daar belangstelling is in wat die spreker sê. Daar is aangetoon dat 'n reaksie vir sprekers die geleentheid gee om hulle gevoelens te lug. Laastens, gee Hoofstuk 6 die gevolgtrekkings van die ondersoek, en die belangrikste bevindinge van die studie word opgesom.
46

Poësie performances : ‘n ondersoek na die moontlikhede vir poësie performance

Borstlap, Mari 03 1900 (has links)
Thesis (MDram)--Stellenbosch University, 2012 / ENGLISH ABSTRACT: Poetry performance as a universal phenomenon is a performing art form. The aim of this study is to explore the diverse nature of this art form. The different types are investigated according to their nature and the reasons provided as motivation for the way in which these types are presented. This study focuses on the following types: poetry found in ritual, poetry readings, poetry recitals, poetry as part of word art (woordkuns), sung poetry, the verse drama and drama’s based on poetry. Under each of these categories relevant examples of each type are discussed. A background study which is primarily focused on a brief overview of the history of poetry performance in relation to its origin and development serves as an introduction and foundation for the study. Hereupon the different types of poetry performances, with relevant and contemporary variations as examples, are being compared to each other with the aim to elucidate the main similarities as well as differences between these examples. As part of the research of poetry performance, a practical project exploring the specific nature of dramas based on poetry was executed. This drama, called Ontslape, shares various comparisons as well as differences with other examples of contemporary drama’s based on poetry. It also shares similarities with word art (woordkuns). Although it is possible to distinguish between certain types of poetry performance, it however also appears that some types overlap in more than one way and therefore the conclusion can be made that poetry performance as a phenomenon consists of various possibilities. / AFRIKAANSE OPSOMMING: Poësie performance as universiële verskynsel is ‘n uitvoerende kunsvorm. In hierdie studie word die diversiteit van die kunsvorm ondersoek. Die onderskeie tipes word ondersoek na aanleiding van die aard en doel van die aanbieding. Die studie fokus op die volgende tipes: poësie in rituele verband, poësie in voorlesing, poësie en voordrag, poësie as deel van woordkuns, poësie in sangvorm, die versdrama en poësie gebasseerde drama’s. Onder elk van hierdie afdelings word relevante voorbeelde van elke tipe breedvoerig bespreek. ‘n Agtergrondstudie wat hoofsaaklik fokus op ‘n historiese oorsig met betrekking tot die oorsprong en ontwikkeling van poësie performance dien as basis vir en inleiding tot die studie. Hierop word onderskeie tipes poësie performance met relevante en kontemporêre voorbeelde en variasies belig met die doel om die ooreenkomste en verskille tussen tipes en voorbeelde uit te lig. As deel van die verkenning van poësie performances is ‘n praktiese projek uitgevoer wat by nader ondersoek ‘n voorbeeld van ‘n poësie gebasseerde drama is. Hierdie drama, getiteld Ontslape, toon verskille en ooreenkomste met ander kontemporêre poësie gebasseerde drama’s. Dit toon ook ooreenkomste met woordkuns. Alhoewel sekere tipes poësie performances duidelik van ander geskei kan word, blyk dit dus dat sekere tipes met mekaar oorvleuel en daar word dus gevolglik tot die konklusie gekom dat poësie performance as verskynsel oor ‘n legio aantal moontlikhede beskik.
47

The influence of gender on the use of hedges by Cantonese speakers in English

Wong, Kuen, Kolya, 黃娟 January 2006 (has links)
published_or_final_version / abstract / Linguistics / Master / Master of Arts
48

Examination of the homophone density effect in spoken Chinese. / CUHK electronic theses & dissertations collection / Digital dissertation consortium

January 2011 (has links)
The results indicate that such effects could be mainly attributed to activation of multiple semantic representations, in line with the primary goal of meaning access for spoken word comprehension. However, it seems that listeners sometimes do activate orthographic information for fine discrimination when the meaning was uncertain. / The study provides the first systematic set of data for the homophone family size effect, demonstrating its presence in spoken Chinese word recognition. / This study examined the homophone family size effect in spoken Chinese using monosyllabic and disyllabic homophones, and disyllabic words with a homophonic syllable. For each of these three types of words, the family size effect was measured in Exp. 1 to 3 using the auditory lexical decision task. In Exp. 1, participants showed slower response to monosyllabic homophones with a larger family size than those with a smaller family size. In Exp. 2, participants showed an opposite family size effect for disyllabic homophones (e.g. /mi4 feng1/[special characters omitted] and [special characters omitted]), that is, higher accuracy to disyllabic homophones than disyllabic non-homophones. In Exp. 3, the family size effect for disyllabic words with a homophonic syllable was found to be dependent on word frequency, an inhibitory effect was present for low frequency words, regardless of whether the first or the second syllable was the homophonic syllable. Using the same three types of words and the same task, Exp. 4 to 6 examined the neural basis of the family size effect with event-related potentials (ERPs). Results showed that the family size effect mainly resulted from access to the multiple semantic meanings associated with a spoken word. In addition, there was evidence that the multiple word-forms of a homophone were activated in a relatively early state (300--500 ms) during spoken word recognition when the first syllable of a disyllabic word was homophonic. / Wang, Wenna. / Advisers: Alan Chun-nang Wong; John Xuexin Zhang. / Source: Dissertation Abstracts International, Volume: 73-07(E), Section: B. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 89-94). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract and appendix also in Chinese.
49

Using Dialogue Acts in dialogue strategy learning : optimising repair strategies

Frampton, Matthew January 2008 (has links)
A Spoken Dialogue System's (SDS's) dialogue strategy specifies which action it will take depending on its representation of the current dialogue context. Designing it by hand involves anticipating how users will interact with the system, and/or repeated testing and refining, and so can be a difficult, time-consuming task. Since SDSs inevitably make understanding errors, a particularly important issue is how to design ``repair strategies'', the parts of the dialogue strategy which attempt to get the dialogue ``back-on-track'' following these errors. To try to produce better dialogue strategies with less time and effort, previous researchers have modelled a dialogue strategy as a sequential decision problem called a Markov Decision Process (MDP), and then applied Reinforcement Learning (RL) algorithms to example training dialogues to generate dialogue strategies automatically. More recent research has used training dialogues conducted with simulated rather than real users and learned which action to take in all dialogue contexts, (a ``full'' as opposed to a ``partial'' dialogue strategy) - simulated users allow more training dialogues to be generated, and the exploration of new dialogue contexts not present in an original dataset. As yet however, limited insight has been provided as to which dialogue contextual features are important to include in the MDP and why. Indeed, a full dialogue strategy has not been learned from training dialogues with a realistic probabilistic user simulation derived from real user data, and then shown to work well with real users. This thesis investigates the value of adding new linguistically-motivated contextual features to the MDP when using RL to learn full dialogue strategies for SDSs. These new features are recent Dialogue Acts (DAs). DAs indicate the role or intention of an utterance in a dialogue e.g. ``provide-information'', an utterance being a complete unit of a speaker's speech, often bounded by silence. An accurate probabilistic user simulation learned from real user data is used for generating training dialogues, and the recent DAs are shown to improve performance in testing in simulation and with real users. With real users, performance is also better than other competing learned and hand-crafted strategies. Analysis of the strategies, and further simulation experiments show how the DAs improve performance through better repair strategies. The main findings are expected to apply to SDSs in general - indeed our strategies are learned and tested on real users in different domains, (flight-booking versus tourist information). Comparisons are also made to recent research which focuses on handling understanding errors in SDSs, but which does not use RL or user simulations.
50

Exploring the Linguistic and the Discourse-pragmatic Functions of Arabic Yaʕni in a Novel Context of Language Use

January 2018 (has links)
abstract: Yaʕni ‘lit. he/it signifies/means/intends’ is an arising linguistic and discourse-pragmatic phenomenon in many varieties and speech situations of spoken Arabic. Yet, the few scholarly investigations yaʕni has received come from restricted and limited contexts of language use. The primary aims of this dissertation were to, first, expand and broaden research on Arabic yaʕni into novel contexts of language use and to, second, explore the linguistic and the discourse-pragmatic functions of yaʕni. Therefore, the data used for this dissertation were collected, selected, and analyzed from a sample of spoken data brought from two episodes of a Saudi sports TV show Alkurah Tatakallam ‘lit. the ball speaks.’ The analytical procedures and discussions showed that yaʕni had the following types of linguistic and discourse-pragmatic functions: as (a) a verb, (b) elaboration and turn expansion, (c) repair organization, (d) managing the turn-taking system, (e) alleviation and hedging, (f) marking concessive/contrastive relations, and (g) emphatic yaʕni. The discussions seemed to suggest the gradual solidification of three views: First, there is a suggestion that the categorical status ranging from verb yaʕni to the discourse marker yaʕni can be understood in terms of scalarity, gradience, and prototypicality. Second, there is another suggestion that gradations can also be located between the discourse-pragmatic functions of yaʕni. Third, there is a suggestion that, synchronically and diachronically, yaʕni as a form has been wildly drifting from its categorical verb status, lexical source, propositional meaning, and even its discourse-pragmatic markerhood. The analysis, discussions, and suggestions invoked the idea of bridging context(s) related to the categorical status and the discourse-pragmatic functions of yaʕni. This categorical status of yaʕni puts the binary distinction between conceptual meaning and procedural meaning of relevance theory, and the studies of yaʕni following such a binary distinction, into question since this distinction seemed blurry. The bridging context(s) seemed to support the gradualness and the directionality of the evolution of DMs. Therefore, the categorical and discourse-pragmatic behavior of yaʕni seems to have support from the hypothesis and theories such as grammaticalization and pragmaticalization. It seems also that the historical development of yaʕni can be discussed in terms of the hypothesis and theories of idiomaticization and phraseology. / Dissertation/Thesis / Doctoral Dissertation Linguistics and Applied Linguistics 2018

Page generated in 0.0205 seconds