Submitted by Diogo Barreiros (diogo.barreiros@ufba.br) on 2017-06-08T11:28:18Z No. of bitstreams: 1 Dissertação Mydiã_Freitas.pdf: 3038925 bytes, checksum: 799df53dc64e5b455ed52bb760695ae2 (MD5) / Approved for entry into archive by Vanessa Reis (vanessa.jamile@ufba.br) on 2017-06-12T15:16:18Z (GMT) No. of bitstreams: 1 Dissertação Mydiã_Freitas.pdf: 3038925 bytes, checksum: 799df53dc64e5b455ed52bb760695ae2 (MD5) / Made available in DSpace on 2017-06-12T15:16:18Z (GMT). No. of bitstreams: 1 Dissertação Mydiã_Freitas.pdf: 3038925 bytes, checksum: 799df53dc64e5b455ed52bb760695ae2 (MD5) / Code smell é definido como um potencial problema de design localizado em determinada parte do código fonte de um sistema de software e pode indicar deficiências no design que podem influenciar a facilidade de manutenção, facilidade de teste e confiabilidade do software. Um bom exemplo de code smell que vai de encontro aos princípios do bom design definido na programação orientada a objetos é o Shotgun Surgery, que corresponde a classes cuja modificação implica em muitas pequenas alterações em muitas outras classes. Quando as modificações estão espalhadas, elas são difíceis de serem encontradas. Como resultado, é fácil perder uma mudança importante, causando assim, problemas na manutenção do sistema. A atividade de identificação de code smells é analisada e confirmada pelo próprio programador ou um revisor independente, contudo, a detecção manual de code smells por meio de inspeções de código consume muito tempo e é pouco escalável. Ferramentas que auxiliam a identificação de code smells têm sido bastante utilizadas em diversos estudos. Neste contexto, esta dissertação propõe investigar se a detecção automática de code smells por meio de ferramentas é considerada útil pelos desenvolvedores como forma de avaliar a qualidade do design de sistemas de software, principalmente em termos de facilidade de manutenção. Realizamos três estudos: sendo um estudo preliminar e dois estudos experimentais similares. O objetivo do estudo preliminar era nos auxiliar a perceber e corrigir problemas que pudessem surgir no decorrer do desenvolvimento dos estudos experimentais. O objetivo desses estudos era investigar a qualidade do design de sistemas de softwares e code smells através do uso de ferramentas de detecção. Os resultados desses estudos indicam que existe relação entre os code smells God Class, Data Class, Shotgun Surgery, Refused Bequest e Feature Envy com problemas de design associados a dificuldades no processo de manutenção de sistemas de softwares.

Researchers from Jean Piaget (1932) to Sehlauoi (2008) have utilized small case studies to intensively examine naturally occurring child language production in the home environment. This qualitative, 5-year longitudinal, five-year case study continues the focus of previous research on child bilingualism and code-switching; however, it aims to strengthen the previously tenuous connection linking said bilingualism with the speech of fraternal boy/girl twins, who are simultaneously making code choices and reflecting their gender language usage of these choices. The purpose of this study was to identify each twin’s code-switching patterns in Spanish and English in the bilingual home environment, examine the differences in their language choices, and reflect on their gendered language use as it pertains to each child’s own language production choices. The study connects the results of this research to previous research on child code-switching and reveals the similarities and differences between these two patterns and new ways of understanding the rich variety of linguistic understandings and possibilities in bilingual speech of twins. The study demonstrates that while Spanish code-maintenance was the preferred code choice by both twins throughout the length of the study, the female twin consistently code-switched into English and took more total turns in English than her twin brother while the male twin code-switched more frequently within a turn. These code-switching occurrences may be related to their gender. The limitations of this study, future directions of research, and conclusions are also discussed.

This study investigated the use of intrasentential code switching on Facebook. The corpus included 1000 screenshots of Facebook comments collected from 10 Saudi female Facebook friends who were bilingual in Arabic and English. The data were examined through statistical and content analysis. The results showed that intrasentential code switching occurs frequently in informal Facebook interactions. Further, the occurrence of code-switching was analyzed in relation to 10 topics of Facebook interactions, including gossip, humor, technology, compliments and thanking, achievement, movies and songs, family and intimacy, makeup, travelling, and religion. Among these 10 topics, gossip and humor elicited significantly higher frequencies of intrasentential code-switching. Moreover, the qualitative results showed that the most frequent English words within Arabic sentences were technical and academic terms, whereas the most frequent Arabic words within English sentences were religious words. Overall, this study shows that the use of intrasentential code-switching among Arabic-English female friends on the social network Facebook is a natural part of their interactions and the frequency with which they employ code-switching is related to the topic of their communication, their language environment, their cultural experiences, and their religion.

In this dissertation, three multiuser detectors are developed for different application scenarios in direct-sequence code-division multiple access systems. The first detector is an overlapping widow decorrelating detector aimed at asynchronous reverse links. In companion with the design of this detector, a study on the decay property of the ideal decorrelating impulse response is presented, resulting in a quantitative description of the decay rate as a function of the Cholesky factors of the cross-correlation matrix of user signature signals. This result can serve as a guide for determining window length of decorrelating or minimum mean-squared error multiuser detection in asynchronous multiuser systems. Based on this result, a signal-adapted window-length determination algorithm is developed for the proposed detector. Several supporting utilities for efficient implementation of the proposed detector are also described. The second detector is a linear multiuser detector that is also aimed at the reverse links. Particularly, it is desirable for cases where the number of users is small and, thus, significant performance gain over the existing linear multiuser detectors is possible. Unlike in the decorrelating and MMSE detectors, minimizing the bit-error rate is taken as the optimization objective in the proposed detector. To avoid undesired local minima of the highly nonlinear BER cost function, a set of convex constraints is proposed for the optimization problem. It is shown that this constrained optimization problem has a unique solution once the decorrelating detector exists. It is also shown that the proposed detector achieves the best performance among linear detectors for most realistic situations. In addition, a Newton barrier method is developed for efficiently calculating the coefficient vector of the proposed detector (i.e., the solution of the constrained optimization problem). The third detector is an adaptive detector that is aimed at the forward link where information about interfering users is often unavailable. The proposed detector consists of a bank of blind adaptive filters, one for each resolvable path, followed by a channel estimator and a coherent diversity combiner. To allow blind adaptation, the impulse response of each filter is decomposed into two orthogonal parts: one part is fixed as the decorrelating coefficient vector for the path in the absence of interfering users and the other is free to be adapted according to the mean-squared error criterion. Assuming perfect adaptation, the performance of the proposed detector is shown to be between those of the decorrelating detector and the minimum mean-squared error detector. Other studies conducted include the effects of fading on the performance of the proposed detector and the behavior of the proposed blind adaptation algorithm. / Graduate

abstract: Corrections has a rich history centered around rehabilitation and its obtainability, and has seen the pendulum swing back and forth between rehabilitative and punitive policies. Currently, there is an emphasis on evidence-based practices which provides a unique opportunity to assess gaps in the rehabilitation literature as a means to ensure that rehabilitative-oriented policies are part of the forefront of corrections. One notable gap in the corrections and rehabilitation literature is that research has not assessed what influences meaningful participation in rehabilitative programming during incarceration. Past research has acknowledged that there is an inmate code, characterized heavily by hypermasculinity, that negatively influences behavior during incarceration, yet research has not examined whether this code influences engagement in rehabilitative programming. The current study seeks to address this gap by examining the inmate code, specifically hypermasculinity, as a barrier to rehabilitation during incarceration through in-depth interviews with five incarcerated individuals from a large Southwestern correctional facility. Findings, limitations, and future research suggestions are discussed. / Dissertation/Thesis / Masters Thesis Criminology and Criminal Justice 2017

Chinese-English code switching (CS thereafter) has become a common phenomenon in mainland China in the last decades of years. Most research focuses on sociolinguistic perspectives of CS in mainland China. However, there is rarely research aiming at exploring linguistic patterns of CS. To fill the research gap, this study attempts to perceive into the linguistic patterns of Chinese-English CS. Myers-Scotton' s Matrix Language Frame model (MLF model thereafter) is used as a framework in this study. On one hand, it can help me to classify data and explain why the data comes into being. One the other hand, CS between two typologically different languages can test the universality of the MLF model. Two hundred and seven Chinese-English bilingual utterances are quantitatively classified and qualitatively described as the basis of the analysis of this study. In this study, a whole picture of linguistic patterns of Chinese-English CS was presented. I further analyze some counterexamples against the MLF model, such as creative forms, which conform to neither the grammar of Chinese nor that of English. Thus, it is revealed that the MLF model cannot offer explanations to those innovative forms. I argue that the innovative forms in the corpus of this study belong to artistic CS. Artistic code­switching utterances mainly exist as lyrics or buzzwords on the Internet, which are intentionally created by people. These forms are invented to be different and attract people,s attention, so usually they do not conform to grammars and common language codes. No wonder that the MLF model cannot explin the artistic CS. Other limitations of the MLF model are also discussed in this dissertation.

abstract: The subject of bilingual language use in the southwestern United States has been widely researched. However research pertaining to the Phoenix Metropolitan area is lacking. Studies have shown that language choice is governed by linguistic as well as social constraints (Gumperz, 1977; Poplack 1980; 1981). This study examined and compared the code-switching behaviors of two communities in the southwestern United States: Antonito, Colorado and the Phoenix Metropolitan area in Arizona. The study explored the social and linguistic factors that are said to govern code-switching behaviors such as the type of switches made (intra-sentential or single lexical switches), the position in the utterance where the switch occurs (final or other), the direction of the switch (an utterance beginning in English and ending in Spanish, or beginning in Spanish and ending in English), the gender and level of education of the participants (college or above; high school or below), the ethnicity of the interviewer (Anglo or Hispanic), as well as which of the aforementioned social and linguistic factors most favored intra-sentential switches. The study used corpus data, with four participants from each community for a total of eight interviews. Participants from each corpus were selected to control for gender, the highest level of education achieved and the ethnicity of the interviewer. The study found that in the corpora looked at, linguistic factors such as position of the switch and direction of the switch affected intra-sentential switches more than social factors, although in terms of frequencies within individual factor groups, social factors such as the ethnicity of the interviewer, and the participant's level of education were found to be significant in affecting code-switching behavior. / Dissertation/Thesis / M.A. Spanish 2013

Submitted by Chaylane Marques (chaylane.marques@ufpe.br) on 2015-03-06T18:08:49Z No. of bitstreams: 2 DISSERTAÇÃO Marcelo da Costa.pdf: 944352 bytes, checksum: d7bff1f647499c4841b1f19f18f05659 (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-06T18:08:49Z (GMT). No. of bitstreams: 2 DISSERTAÇÃO Marcelo da Costa.pdf: 944352 bytes, checksum: d7bff1f647499c4841b1f19f18f05659 (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2013-03-01 / Nas aulas de língua estrangeira, sobretudo nas que adotam perspectivas mais comunicativas, é comum a alternância de uso das línguas materna e estrangeira. No entanto, pouco sabemos sobre o fenômeno de mudança de código (code-switching) nesses contextos, já que muitos dos estudos sobre o tema concentram-se em situações naturais de bilinguismo e em países em que duas ou mais línguas são corriqueiramente faladas pela comunidade. Questões como “por que professor e alunos mudam de uma língua para outra na interação em sala de aula de língua estrangeira?”, “há pontos de ocorrência específicos prioritários na alternância?”, “que papel tem outras variáveis tais como nível do curso (básico e avançado), autoria e tipos na alternância de códigos?” poderiam fornecer subsídios valiosos para o ensino. Esta dissertação propõe uma análise das mudanças de código (code-switching) realizadas por ambos alunos e professores e quais as consequências do seu uso na interação e no ensino-aprendizagem de uma língua estrangeira. Este trabalho ainda busca desenvolver relações com os pontos de ocorrência dessas mudanças de língua e analisar se estas são prefaciadas por marcadores conversacionais ou expressões específicas. A pesquisa é etnográfica de caráter qualitativo e quantitativo. A coleta dos dados foi realizada no Núcleo de Línguas e Cultura da Universidade Federal de Pernambuco (NLC – UFPE). As duas turmas observadas são de nível Básico 1 e Avançado 1, totalizando 20 horas-aula de observação. Os resultados apontam que existem mudanças exclusivas quanto ao nível do aluno e também sugere que apesar do foco ser em falar o idioma alvo, o professor, que deveria liderar a interação, realiza mais mudanças que o aluno.

[pt] Neste trabalho são examinadas estruturas CELP com baixo retardo (da ordem de 2 ms) para codificação digital de voz a taxas de bits de 8 k, 9,6 k e 16 kbit/s. São avaliados sete algoritmos de predição adaptativa a partir do sinal sintetizado. Esses algoritmos são usados para adaptação do filtro de predição com retardo curto e do filtro de ponderação de codificadores CELP com baixo retardo. Dos algoritmos analisados, um emprega adaptação em bloco e os outros são recursivos utilizando estruturas transversais e em treliça. As comparações são baseadas em termos das medidas objetivas de razão sinal ruído (RSR), razão sinal ruído segmentar (RSRSEG) e razão sinal ruído ponderada em freqüência (RSRW) e em termos de medidas subjetivas obtidas através de testes de escuta informais. / [en] Low delay (about 2 ms) CELP structures are studied for digital coding of speech at 8, 9,6 and 16 kbit/s. Seven backward adaptative prediction algorithms are investigated. These algorithms are used for adapting the short-term prediction filter and the weighting filter in low delay CELP coders. One algorithm uses block adaptation and the others are recursive with transversal and lattice structures. Comparisons are based on the objective measures of signal to noise ratio, segmental signal to noise ratio and a frequency weighted signal to noise and on subjective results obtained from informal listening tests.

The initial code Acquisition Techniques of Direct Sequence Spread Spectrum Systems for two categories of serial and parallel search strategies is investigated. A simple and economic scheme for coarse code acquisition of Reverse Link for UMTS (FDD-WCDMA) application is presented. The emphasis is on the performance of a new scheme (using a Surface Acoustic Wave Matched Filter) as a term of probability of false detection (pf) in code division multiple access (CDMA) systems. Knowledge of initial code uncertainty phases help to reduce the overhead of preamble on the access channel and a very simple scheme for acquisition to be determined. In the reverse link this uncertainty is due to the cell radius only. Acquisition time required for a simple serial search scheme may therefore be unacceptably large. On the other hand, for parallel acquisition using parallel branches in accordance with the chip uncertainty time region leads to a lot of hardware complexity. Initially, the effect of multiple access interference and spreading sequence length are determined for models applicable to the reverse link of a mobile communications system. Then the acquisition performance is derived using a model of a cellular mobile communication channel, which includes the effects of multiple access interference, adjacent cell interference, frequency selective Rician channel, shadowing, power control error, and vehicle speed. It is shown that the most significant factors in determining the acquisition performance are the acquisition observation interval, the number of users, and the specular to diffuse power ratio. Numerical results based on analysis of acquisition performance in mobile channel show that the proposed acquisition scheme is efficient, robust, fast and suitable for real time low cost implementation.