Browsing by Author "Alotaibi, Amal"

Now showing 1 - 2 of 2

Restricted
Perception of Regional Spoken Arabic by Native Speakers
(George Mason University, 2024-05) Alotaibi, Amal; Lukyanenko, Cynthia
This dissertation examines native speakers’ word recognition of, differentiation between, and social attitudes toward varieties of Arabic. It is a particularly interesting test case because of the Arabic unique regional variation situation and the available literature lacks data on how Arabic speakers perceive different accents, with a particular emphasis on their connections to the speakers' sociological and regional backgrounds. Therefore, the main purpose of this study is to discover how native speakers perceive various Arabic speech varieties and test the accent familiarity effect to determine the effects of dialectal variation and language experience on speech perception. Specifically, whether the availability of information regarding the speakers' accent in the speech signal would influence the recognition of spoken words and sentences. To do this, I examine how two groups of native Arabic speakers; Najdi Arabic (NA) and Saudi Southern Arabic (SA), perceive and adapt to three different regional accents; NA, SA (‘own’ or ‘nearby’ accent), and Egyptian Arabic (EA) (‘distant’ accent). I conducted three perception studies to explore NA and SA speakers’ processing of regional Arabic varieties. In the first experiment, I examined participants' ability to recognize speech stimuli in their ‘own’, a ‘nearby’, or a ‘distant’ variety. NA and SA participants were asked to make a lexical decision (‘word’ or ‘nonword’) on target items placed at the end of sentences spoken by NA speaker, SA speaker, and EA speaker. Results show that participants were good at recognizing ‘words’ from ‘nonwords’ with an accuracy level of (93.3%). Moreover, ‘nonword’ trials have slightly slower reaction times compared to the ‘word’ trial type, especially for the ‘distant’ accent since it is not that familiar to them. Similarly, SA participants’ performance in ‘nonword’ trials shows slower reaction times as compared to the performance of NA participants. This demonstrates how regional accents can affect word recognition and that responding to a ‘distant’ variety requires more time and effort from the listeners. In the second experiment, I examined participants' ability to distinguish between the different regional accents. Another set of NA and SA participants performed a discrimination task where they were asked to determine whether two different talkers were from the ‘same’ region or ‘different’ regions. Results from this study show that all participants had relatively similar reaction times. In terms of trial types, responses from 'different' trials had faster reaction times, particularly those with 'distant' dialect (where EA is one of the combinations of the two audio samples). In the third experiment, I examined participants' attitudes, social representations, and social judgments toward the same regional accents, NA, SA, and EA. A new group of NA and SA participants were asked to rate nine audio samples spoken by three NA speakers, three SA speakers, and three EA speakers on social and personal traits, including accentedness, on a 6-point rating scale. Results from this social judgment task reveal that participants from both groups were lenient with speakers who speak their ‘own’ variety, especially in accentedness ratings. The statistical analyses also reveal significant main effects of participant accent and talker accent across multiple characteristics. Taken together, the findings of these three studies have shed light on the effects of familiarity with the own Arabic variety (familiar accent), nearby Arabic variety (less familiar accent), and distant Arabic variety (unfamiliar accent) on accents’ perception and recognition. In particular, the present research provides us with a better understanding of how native Arabic speakers generally handle the linguistic variation they encounter in speech in their daily life, recognize regional accented words, distinguish between regional accents, and express their own social views and accent ratings toward these various regional accents that are either ‘own’, ‘nearby’, or ‘distant’ accent to the participants. It will contribute to our comprehension of how accent perception works in general, how native Arabic speakers recognize regionally accented words and nonwords, discriminate between different regional accents, and evaluate the sociological background of regionally accented talkers.
20 0
Restricted
Semantic Analysis of Amazon Reviews of Sustainable Products
(University of Leeds, 2024-02-18) Alotaibi, Amal; Dimitrova, Vania
Online shopping has grown to be an essential part of modern living, garnering a wealth of client input. This project advances the field of consumer feedback mining and semantic and sentiment analysis of customer reviews since, when applied effectively, it can enhance goods, services, or marketing initiatives. This project proposes a framework using Natural Language Processing (NLP) techniques to find customer preferences related to sustainability through mining customer reviews (CR) text. First, implement the LDA and sLDA models using the Gensim package in Python to extract sustainable topics from CR. After that, implement the BERTopic model to find the sustainability aspect in (CR). Then, the overall sentiment for every review in each topic was calculated using the Vader sentiment library in Python. Lastly, interpret the results and generate helpful insights for brand managers. The Amazon product review data is used in this study, and we use Food and Grocery Sustainable Products. The findings of the proposed framework are promising, as we were able to identify the most discussed topics in sustainability aspects of products and produce an assessment that provides information about the aspects that the customers are most satisfied with and that can be improved. However, the sLDA model and the BERTopic model achieve the goal but not the expectation. especially BERTopic, it was not accurate enough for weakly supervised text classification. Also, the Vader sentiment tool did not meet expectations because of the complexity of CR. However, the text analyst specialist found that the structure is flexible enough to allow for future development and increased usage. Ultimately, we think that these data will help brand managers create and improve future products, which will raise consumer satisfaction and boost revenue and profitability.
25 0