SABA Publishing
Not a member yet
602 research outputs found
Sort by
YorubaAI: Bridging Language Barrier with Advanced Language Models
Full text linkYorubaAI addresses the digital divide caused by language barriers, particularly for Yoruba language speakers who struggle to interact with advanced large language models (LLMs) like GPT-4, which primarily support high-resource languages. This study develops a system, named YorubaAI, for seamless communication in Yoruba language with LLMs. The YorubaAI enables users to input and receive responses in Yoruba language, both in text and audio formats. To achieve this, a speech-to-text (STT) model is fine-tuned for automatic Yoruba language speech recognition while a text-to-speech (TTS) model is employed for conversion of Yoruba language text to speech equivalent. Direct communication with LLM in low-resource languages like Yoruba language typically yields poor results. To prevent this, a generation technique known as retrieval-augmented generation (RAG) is utilized to augment the LLM's existing knowledge with additional information. The RAG is formed through creation of a database of questions and answers in Yoruba language. This database serves as the primary knowledge base that the YorubaAI uses to retrieve relevant information with respect to the question asked. The content of the created questions and answers database is converted into vector embeddings using Google’s Language-Agnostic BERT Sentence Embedding (LaBSE) model to yield numerical representations that capture the semantic meaning of the texts. The embeddings generated from the Yoruba questions database are stored in a vector store database. These embeddings were essential for efficient search and retrieval.The the two models (STT and TTS models) were integrated with a LLM using a user-friendly interface that was built using the Gradio framework. The STT model achieved a word error rate of 13.06% while the TTS model generated natural-sounding Yoruba language speech. YorubaAI correctly responded to various queries in pure Yoruba language syntax and thus successfully bridges the AI accessibility gap for Yoruba language speakers
The Impact of Quality of Work Life on Job involvement at Yemen Mobile Telecommunication Company – Republic of Yemen
Full text linkهدفت الدراسة إلى التعرف على أثر جودة الحياة الوظيفية بأبعادها (الأُجور والمكافآت، والمشاركة في اتخاذ القرارات، وبيئة العمل، والاستقرار والأمان الوظيفي) في الاستغراق الوظيفي في شركة يمن موبايل للهاتف النقال، وقياس مستوى جودة الحياة الوظيفية وكذا قياس واقع الاستغراق الوظيفي لدى موظفي شركة يمن موبايل، وقد استخدم المنهج الوصفي التحليلي، والاستبانة أداةً لجمع البيانات
واستهدفت الدراسة جميع العاملين في شركة يمن موبايل البالغ عددهم (543) موظفاً وموظفة، وبلغ حجم العينة العشوائية البسيطة (226) موظفاً وموظفة، وتم استرداد (195) استبانة صالحة، وبنسبة 86% من حجم العينة. وقد خلصت الدراسة إلى العديد من الاستنتاجات، كان أبرزها: أن مستوى جودة الحياة الوظيفية في شركة يمن موبايل للهاتف النقال جاء بمستوى متوسط، وأن مستوى الاستغراق الوظيفي في شركة يمن موبايل جاء مرتفعًا، وأظهرت نتائج الدراسة وجود أثر لجودة الحياة الوظيفية بأبعادها (الأجور والمكافآت ، والمشاركة في اتخاذ القرارات، وبيئة العمل، والاستقرار والأمان الوظيفي) في شركة يمن موبايل للهاتف النقال في الاستغراق الوظيفي للعاملين. وخلصت الدراسة إلى توصيات عدة، أهمها: تُوصى بزيادة الاهتمام بتعزيز جودة الحياة الوظيفية للعاملين. وإعادة النظر في سياسات الأجور والمكافآت، وبالحفاظ على مستوى الاستغراق الوظيفي المرتفع وتحسين نظام تأمين على الحياة ومعاش تقاعدي جيد، وإعادة هيكلة نظام الأجور والمكافآت.The study aimed to examine the impact of Quality of Work Life (QWL)—represented by its dimensions (salaries and rewards, participation in decision-making, work environment, and job security and stability)—on job engagement at Yemen Mobile Company for Mobile Telecommunication. It also sought to measure the overall level of QWL and assess the current state of job engagement among the company’s employees.
The researcher adopted the descriptive–analytical method, using a questionnaire as the primary tool for data collection. The study targeted all employees of Yemen Mobile, whose total number was 543 male and female employees. A simple random sample of 226 employees was selected, and 195 valid questionnaires were retrieved, representing 86% of the total sample size. The study reached several conclusions, the most significant of which are as follows: the level of Quality of Work Life at Yemen Mobile Company was found to be moderate, while the level of job engagement among employees was high. Moreover, the results revealed that Quality of Work Life and its dimensions (salaries and rewards, participation in decision-making, work environment, and job security and stability) have a significant effect on job engagement among employees at Yemen Mobile. Based on these findings, the study recommended increasing attention toward enhancing Quality of Work Life for employees, reconsidering salary and reward policies, maintaining the high level of job engagement, improving the life insurance and retirement systems, and restructuring the salary and incentive framework to ensure fairness and motivation
A Fluctuation Analysis of United State Dollar Price in the Philippines: A Stochastic White Noise Approach
Full text linkGiven the fluctuating trend in the United States Dollar (USD) exchange rate in the Philippines, a stochas-tic analysis employing a white noise framework was conducted to examine the daily USD price. This approach was utilized to characterize the statistical behavior of the exchange rate by computing the Mean Square Deviation (MSD) based on 11,205 daily data points spanning from January 1978 to July 2023. The resulting theoretical MSD was used to generate a Probability Density Function (PDF), which effectively describes the empirical distribution of the USD price fluctuations in the Philippine context
Metacognitive awareness and language achievement test performance among Iranian undergraduate ESP students
Full text linkMetacognitive awareness plays a fundamental role in enabling learners to employ appropriate strategies in problem solving, and monitoring their learning process in different conditions. The current study aims at finding the metacognitive awareness level of undergraduate students in ESP. To this end, 33 undergraduate students in ESP took part in the study. They were asked to respond to the given statements provided in the inventory. The data were gathered and analyzed. The results indicated that the participants' metacognitive awareness is high. Moreover, the participants obtained high mean scores on the two components of metacognitive awareness. The findings can be useful to learners to become familiar with their level of metacognitive awareness and to use them properly in different conditions. The findings can also be conducive to instructors and material developers to deliver materials in such a way that best increases learners' metacognitive awareness and involvement in learning process
The Effectiveness of corpus based listening sessions for intermediate language learners
Full text linkCorpus‑based approaches reposition authentic spoken data as the primary resource for listening instruction, arguing that learners develop stronger active listening skills when they encounter real usage patterns rather than contrived textbook examples. By making recurrent lexical bundles, discourse markers, prosodic tendencies, and interactional routines visible, corpora enable teachers and learners to focus on the forms and functions that actually shape comprehension in natural speech. This orientation reduces the artificial gap between classroom input and real‑world listening demands and foregrounds the processes—prediction, noticing, mapping form to function, and strategic response—that underpin effective active listening
Ensemble-based Intrusion Detection System for Electric Vehicles Charging Stations using Machine Learning
Full text linkTraditional Vehicles have an adverse effect on the environment. Therefore, the current technological shift is constantly seeking an alternative to replace traditional vehicles fueled by fossil fuels, and Electric vehicles are, so far, the best alternative. The adoption of Electric Vehicles (EVs) is growing rapidly due to their eco-friendly benefits and technological advancements. This growth, however, brings a significantly larger attack surface due to increased interconnectivity between electric vehicles, charging stations and the smart grid system. To prevent such types of attacks, we need a robust system to detect them beforehand and prevent the system from being compromised. Although some prior work has been conducted in this area, their approaches did not incorporate deep learning algorithms, nor did they evaluate model performance under noisy data conditions. Therefore, we proposed a novel ensemble-based intrusion detection system (IDS) to detect these attacks in Electric Vehicle Charging Stations (EVCS). We implement different Machine learning algorithms such as k-nearest neighbors (KNN), Logistic Regression (LR), Support Vector Machine (SVM) and Decision Tree (DT). Moreover, as different types of malwares often exhibit distinct structural characteristics when visualized as images, we also use Convolutional Neural Networks (CNNs) to detect such attacks and malware. We are focusing on detecting attacks in Electric vehicle charging stations by analyzing the network traffic. For this, we utilize the latest labelled dataset, the Canadian Institute of Cybersecurity EV Charger Attack Dataset 2024 (CICEVSE2024), which is a multidimensional dataset containing both benign and attack data. We then evaluate & compare the performance of these algorithm in detecting the network traffic attacks in Electric Vehicle Charging Stations (EVCS). Our proposed model employs an ensemble voting strategy to combine the predictions from different classifiers, thereby improving the system's robustness and accuracy, and achieves an accuracy of 99.5% in detecting cyberattacks. With the addition of small noise to the dataset, a few individual classifiers perform poorly; however, the ensemble model still maintains an accuracy of 99.2%
DCGAN Beyond Generation: A Critical Review of The Performance and Challenges of Forensic Face Models
No full textComputer vision and deep learning techniques, especially deep convolutional generative adversarial networks (DCGAN), have enabled advanced mechanisms to address complex challenges in forensics, especially in reactivating cold case investigations. Cold cases present unresolved challenges due to deteriorating or scarce visual evidence. This paper provides a systematic review that analyzes, classifies, and evaluates the current status of DCGAN and related GAN structures in legitimate face modeling. The primary objectives are to evaluate reported methodologies, performance metrics, and limitations across key applications, including sketch-to-image conversion. The review identifies significant methodological gaps, particularly the absence of standardized assessment measures and the critical challenge of identity preservation. Furthermore, the research explores the ethical and legal considerations associated with computer-generated facial images, focusing on algorithmic bias, accountability, and legal admissibility in criminal investigations. The paper concludes by highlighting key research gaps and proposing future directions necessary for the robust, reliable, and ethically responsible deployment of GAN systems in legitimate practic
Fractional Powersets and SuperHyperStructures: Toward a Framework for Fractional Set Theory and Discrete Hierarchical Systems
Full text linkHyperstructures build on powersets to model multivalued relations on a base set; SuperHyperstructures iterate the powerset to capture layered hierarchies and richer composition. Prior work typically fixes the iteration height to a nonnegative integer. This paper asks whether fractional, inverse, and complex (including imaginary) “heights" can be incorporated coherently. We introduce the notions of an m-root powerset (peeling a specified number of subset layers), a negative powerset (a partial inverse of iterated powersets under a given presentation), and a complex-height powerset defined at the level of observables via operator-theoretic interpolation. We characterize when these operators are well defined—by exponential-tower size conditions in the finite case and by the beth hierarchy in the infinite case—and establish exact inverse laws on their natural domains.Lifting from carriers to operations, we obtain root and negative SuperHyperStructures that preserve incidence, compose naturally, and recover the original structures after the appropriate number of lifts. Conceptually, the framework provides a principled, continuous interpolation across hierarchical levels and a reversible mechanism for descending them, suggesting applications to discrete modeling, policy design, and multi-resolution analysis
The Rhetorical Conflict in Greer and Farzagh: (Study in Dialogue and Text Interaction)
Full text linkيتناول هذا البحث آليات الصراع الخطابي في نقائض جرير والفرزدق، من خلال مقاربة حوارية وتفاعلية للنصوص الشعرية المتبادلة بين الشاعرين. ويُظهر التحليل أن شعر النقائض لا يُبنى على أحادية الصوت، بل يقوم على مناظرة حجاجية، تُؤسسها التفاعلية النصية بين القصائد، بحيث تنشأ كل قصيدة كرد واعٍ ومباشر على سابقتها، مما يجعل النص الشعري الثاني يتخلق من رحم الأول، ويتجاوب معه ضمن جدلية خطابية قائمة على الادعاء والاعتراض.
وتتجلى في هذا الصراع الخطابي ثنائيات ضدية (أنا/آخر، قوة/ضعف)، حيث يتخذ الشاعر من فخره سلاحًا في ترسيخ الفاعلية الإيجابية، في حين يُوظّف الهجاء لتقويض صورة الخصم وإظهاره في موقع الفاعلية السلبية.
ويخلص البحث إلى أن تمثلات الصراع عبر الحوارية والتفاعلية النصية في نقائض جرير والفرزدق تتجسد عبر آليات متعددة، أبرزها التناظر، والتضاد، والاعتراض، ما يجعل كل نص بمثابة قراءة تفكيكية وتأويلية للنص السابق، تُسهم في إنتاج خطاب جديد مضاد، يحاول تقويض الخطاب الخصمي وإعادة تشكيل دلالاته بما يخدم الموقف الشعري للمنتج الثاني.This research examines the manifestations of conflict in the ruins of Jarir and Farzagq, through a dialogical and interactive approach to the poetic texts exchanged between the two poets. The analysis shows that the poetry of the fallen is not based on monophonic sound, but based on an argumentative debate, established by the textual interaction between poems, so that each poem arises as a conscious and direct response to the previous, which makes the second poetic text creates from the womb of the first, and responds to it within a rhetorical dialectic based on the claim and objection.
This rhetorical struggle is manifested in anti-binaries (me/another, strength/weakness), where the poet takes his pride as a weapon in establishing positive effectiveness, while using satire to undermine the image of the opponent and show him in the position of negative effectiveness.
The research concludes that the representations of conflict through dialogue and textual interaction in the faults of Jarir and Farzadq are embodied through multiple mechanisms, most notably symmetry, opposition, and objection, which makes each text as a deconstructive and interpretive reading of the previous text, contributing to the production of a new counter-discourse, which tries to undermine the adversarial discourse and reshape its connotations to serve the poetic position of the second product
Dr EXPLANATORY ANALYSIS OF GRADUATE EMPLOYABILITY SKILLS, ENTREPRENUERSHIP SELF-EFFICACY AND PERFORMANCE EFFICIENCY DRIVE ON JOB CREATION ABILITY IN SOUTHWESTERN NIGERIA: Unveiling the Paradox of Skills, Efficacy, and Entrepreneurial Outcomes
No full textتعتمد القدرة على خلق فرص العمل على الفلسفة الاجتماعية والاقتصادية التي تؤكد على أن الفرص تُخلق كخطوات استباقية للتصدي للبطالة وآثارها المصاحبة التي أثرت بشكل كبير في حياة وسبل عيش الشعب النيجيري. تبلغ نسبة البطالة حاليًا 34.3٪. ركزت الدراسات السابقة بشكل أكبر على تحليل أسباب وتدخلات التحديات الاقتصادية للبطالة ونقص العمل، بالإضافة إلى إستراتيجيات تصفح الإنترنت للبحث عن وظائف عبر الإنترنت أو عن بُعد، بدلاً من تجنب المتغيرات المحفزة التي يمكن أن تحفز القدرات الإبداعية الوظيفية. وبالتالي، بحثت هذه الدراسة في مهارات قابلية التوظيف للخريجين، والكفاءة الذاتية في ريادة الأعمال، وكفاءة الأداء على القدرة على خلق فرص العمل في جنوب غرب نيجيريا. اعتمدت الدراسة تصميم البحث الوصفي من النوع الارتباطي، مع اعتماد أسلوب أخذ العينات متعدد المراحل. تم إحصاء الولايات الست في الجنوب الغربي، بينما استُخدمت تقنية أخذ العينات العشوائية البسيطة لاختيار ثلاث ولايات هي لاغوس وأوسون وإكيتي. كما استُخدمت تقنيات أخذ العينات العشوائية البسيطة لاختيار 900 خريج من القائمة التي تم الحصول عليها رسميًا من وكالة التعليم الحكومية. من خلال استبيان مُمكّن بالتطبيق، استُخدمت Survey Heart لنشر أدوات، وهي مقياس القدرة الإبداعية للخريجين (r=0.77)، ومقياس تقييم مهارات قابلية التوظيف للخريجين (r=0.81)، ومقياس الكفاءة الذاتية لريادة الأعمال (r=0.72)، ومقياس دافع كفاءة الأداء (r=0.89)، والتي استمرت ثمانية أسابيع. حُلّلت البيانات الكمية باستخدام الإحصاء الوصفي، ومعامل ارتباط بيرسون عند مستوى دلالة 0.05. وجدت الدراسة أن قدرة الخريجين على خلق فرص العمل كانت مرتبطة سلبًا بمهارات قابلية التوظيف، ودافع كفاءة الأداء، بينما لم تُظهر الكفاءة الذاتية لريادة الأعمال أي تأثير يُذكر× مما يُبرز عدم التوافق بين المهارات المكتسبة ونتائج ريادة الأعمال. وبشكل مُجتمع، ساهمت العوامل المُتنبئة مساهمة متواضعة، ولكنها مهمة (3.5%) في قدرة خلق فرص العمل؛ مما يُشير إلى أن العوامل الهيكلية والسياقية الأوسع تلعب دورًا أقوى، وقد تم تطبيق النتائج والتوصيات وفقًا لذلك.Job creation ability rests on the socioeconomic philosophy which affirmed that opportunities are created as proactive steps to wade off unemployment and its attendant effects which have largely affected the lives and livelihoods of the Nigerian people. Currently, the unemployment statistics stand at 34.3 %. Previous studies focused more on the analysis of the causes and interventions to economic challenges of unemployment and underemployment as well as strategies to surf the internet for online or remote jobs establish than averting precipitating variables that can trigger job creative abilities. Thus, the study investigated graduate employability skills, entrepreneurship self-efficacy and performance efficiency on job creation ability in southwestern Nigeria. The study adopted descriptive research design of the correlational type while multi-stage sampling procedure was adopted. The six states (LGAs) in Southwestern were enumerated while simple random sampling technique was used to select three states namely Lagos, Osun, and Ekiti States. Also, simple random sampling techniques was used to select 900 graduates from the list officially obtained from government education agency. Through, app-enabled survey, Survey Heart was used to deployed instruments namely Graduate Creative Ability Scale (r=0.77), Graduate Employability Skills Rating Scale (r=0.81); Entrepreneurship Self-efficacy Scale (r= 0.72) and Performance Efficiency Drive Scale (r=0.89) which lasted eight weeks. The quantitative data were analysed using descriptive statistics and Pearson product moment correlation at 0.05 level of significance. The study found that graduates’ job creation ability was negatively related to employability skills and performance efficiency drive, while entrepreneurship self-efficacy showed no significant influence, highlighting a mismatch between acquired skills and entrepreneurial outcomes. Collectively, the predictors made a modest yet significant contribution (3.5%) to job creation ability, suggesting that broader structural and contextual factors play a stronger role. The implication of the findings and recommendations were made accordingly