Thesis. 1979. Ph.D.--Massachusetts Institute of Technology. Dept. of Linguistics and Philosophy. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND HUMANITIES. / Vita. / Bibliography: leaves 419-426. / by John Joseph McCarthy, III. / Ph.D.

This thesis discusses different approaches to machine translation (MT) from Dialectal Arabic (DA) to English. These approaches handle the varying stages of Arabic dialects in terms of types of available resources and amounts of training data. The overall theme of this work revolves around building dialectal resources and MT systems or enriching existing ones using the currently available resources (dialectal or standard) in order to quickly and cheaply scale to more dialects without the need to spend years and millions of dollars to create such resources for every dialect. Unlike Modern Standard Arabic (MSA), DA-English parallel corpora is scarcely available for few dialects only. Dialects differ from each other and from MSA in orthography, morphology, phonology, and to some lesser degree syntax. This means that combining all available parallel data, from dialects and MSA, to train DA-to-English statistical machine translation (SMT) systems might not provide the desired results. Similarly, translating dialectal sentences with an SMT system trained on that dialect only is also challenging due to different factors that affect the sentence word choices against that of the SMT training data. Such factors include the level of dialectness (e.g., code switching to MSA versus dialectal training data), topic (sports versus politics), genre (tweets versus newspaper), script (Arabizi versus Arabic), and timespan of test against training. The work we present utilizes any available Arabic resource such as a preprocessing tool or a parallel corpus, whether MSA or DA, to improve DA-to-English translation and expand to more dialects and sub-dialects. The majority of Arabic dialects have no parallel data to English or to any other foreign language. They also have no preprocessing tools such as normalizers, morphological analyzers, or tokenizers. For such dialects, we present an MSA-pivoting approach where DA sentences are translated to MSA first, then the MSA output is translated to English using the wealth of MSA-English parallel data. Since there is virtually no DA-MSA parallel data to train an SMT system, we build a rule-based DA-to-MSA MT system, ELISSA, that uses morpho-syntactic translation rules along with dialect identification and language modeling components. We also present a rule-based approach to quickly and cheaply build a dialectal morphological analyzer, ADAM, which provides ELISSA with dialectal word analyses. Other Arabic dialects have a relatively small-sized DA-English parallel data amounting to a few million words on the DA side. Some of these dialects have dialect-dependent preprocessing tools that can be used to prepare the DA data for SMT systems. We present techniques to generate synthetic parallel data from the available DA-English and MSA- English data. We use this synthetic data to build statistical and hybrid versions of ELISSA as well as improve our rule-based ELISSA-based MSA-pivoting approach. We evaluate our best MSA-pivoting MT pipeline against three direct SMT baselines trained on these three parallel corpora: DA-English data only, MSA-English data only, and the combination of DA-English and MSA-English data. Furthermore, we leverage the use of these four MT systems (the three baselines along with our MSA-pivoting system) in two system combination approaches that benefit from their strengths while avoiding their weaknesses. Finally, we propose an approach to model dialects from monolingual data and limited DA-English parallel data without the need for any language-dependent preprocessing tools. We learn DA preprocessing rules using word embedding and expectation maximization. We test this approach by building a morphological segmentation system and we evaluate its performance on MT against the state-of-the-art dialectal tokenization tool.

Texte remanié de : Thèse doctorat : Linguistique : Grenoble 3, 1991. / Bibliogr. p. 397-416.

Not available / text

This is a psychoacoustic study investigating experimentally the role of intonation as indicative of the human phenomenon of emotion in both Arabic and English. Itstudies both the acoustic properties of emotion in speech and their impact on intonational contours.Utterances representing five emotions (anger, fear, happiness, sadness, and neutral) in both the declarative and interrogative modes were collected from the speech of eight professional actors (4 Arabic, and 4 English) as they performed roles in movies and drama series. Two types of judges were used: viewers and listeners. The former watched the clips carrying the utterances and identified their emotional content. Their responses determined which utterances were included in the acoustic analysis. The listeners listened only to the utterances chosen by the viewers, and their responses were used to determine the acoustic clues for emotions. The acoustic analysis involved measuring the parameters of fundamental frequency (FO), intensity, and duration of four units of analysis: utterance as a whole unit, the initial and the final syllables of the utterance, and the syllable with the highest FO value (the peak).The ANOVA statistical test was run on the acoustic data. The listeners' responses were used in the Kappa test to determine their emotion recognition accuracy.The results showed that no single parameter can be taken as the sole marker or clue to a certain emotion. Rather, the expression of emotion is viewed as a complicated process involving the three parameters combined. Profiles for each emotion involving the levels of the three parameters at both the utterance and syllable levels are provided. The data analysis did not show emotion to have an impact on international contours. The KAPPA test showed a high degree of emotion recognition accuracy in both languages. The comparison of Arabic and English showed differences in the three parameters between the two languages. The most remarkable feature distinguishing the people of the two languages speech is intensity, with Arabic speakers showing higher decibel levels. / Department of English

This study investigated the pragmatics of codeswitching from FuSHa Arabic, the high variety of Arabic (FA), to Aammiyyah Arabic, the low variety or vernacular (AmA), in the most formal type of discourse, namely religious-oriented discourse.The study posited the following five hypotheses:1) CS occurs with considerable frequency in religious discourse; 2) these switches are communicatively purposeful; 3) frequency of CS is related to the linguistic make-up of the audience addressed, 4) to the AmA of the speaker, and 5) to the section of the discourse delivered.To carry out the investigation, the researcher analyzed 18 audio and videotapes of religious discourse, delivered by 13 Arabic religious scholars from different Arab countries. Ten of these tapes were used exclusively to show that CS occurs in religious discourse. The other eight tapes were used to investigate the other hypotheses. The eight tapes involved presentations by three of the most famous religious scholars (from Egypt, Kuwait, and Yemen) delivered 1) within their home countries and 2) outside their home countries.Three of the five hypotheses were supported. It was found that: CS from FA to AmA occurred in religious discourse with considerable frequency; these switches served pragmatic purposes; and the frequency of the switches higher in the question/answer sections than in the lecture sections.Analysis showed that codeswitches fell into three categories: iconic/rhetorical, structural, and other. The switches served numerous communicative functions, some of which resemble the functions found in CS in conversational discourse.One finding was the relationship between the content of the message and the attitude of the speaker toward or its source. Generally, what the speakers perceived as [+positive] was expressed by the H code, and whatever they perceived as [-positive] was expressed by the L code. Scrutiny of this exploitation of the two codes indicated that FA tended to be utilized as a means of upgrading, whereas AmA was used as a means of downgrading. / Department of English

Thesis (Ph.D.) -- University of Western Sydney, 2005. / "A thesis presented to the University of Western Sydney, College of Arts, Education and Social Sciences, School of Languages and Linguistics, in fulfillment of the requirements for the degree of Doctor of Philosophy, 2005." Includes bibliographical references and appendices.

Version remaniée de la th. : linguistique théorique et formelle : Paris 7 : 1995. / Includes bibliographical references (p. [212]-219) and index.

Thesis (Ph. D.)--University of Nijmegen, 1998. / Includes bibliographical references (p. [409]-424).

Teaching Arabic as foreign language is a fairly new field of study, particularly in Western Universities. The aim of this thesis is to discuss one aspect of this field by concentrating on text-book analysis and evaluation. In carrying out this task we have formulated a set of questions to serve as parameters in the collection and analysis of data relevant to this task (appendix II). The subject of this practical study is vol. 1, entitled "Arabic for Beginners" of a series of books used for teaching Arabic to foreign adults at the Arabic Language Institute. Our study utilises a number of notions and considerations which are applicable to text-book analysis and evaluation generally. In Appendix I we deal with some of these points in an attempt to specify the theoretical dimension presupposed by our practical analysis. The thesis contains two main parts:- the theoretical part which deals with different aspects belonging to the teaching of Arabic as a foreign language, including the characteristics of the Arabic language and the teaching of sounds, vocabulary, constructions, and culture. The practical part, on the other hand, is concerned with the practical application and the outcome of the analysis of the text-book in terms of a set of criteria which may have applicability outside the immediate realm of Arabic foreign language teaching. The thesis contains four chapters. Chapter one deals with the characteristics of the Arabic language and its importance as a foreign language; it also concentrates on the objectives of Arabic language teaching as well as on the problem of syllabus design in relation to this language. Chapter two gives a general outline of the institutes and the materials of teaching Arabic as a foreign language in Saudi Arabia. Chapter three deals with the importance of sounds, vocabulary, constructions and culture, paying attention to their place in the text-book of teaching Arabic as a foreign Language. Chapter four sets out the results of applying the criteria of text-book evaluation to the book under investigation, pointing out its positive and negative features in the areas of sounds, vocabulary, constructions and culture.