VFAST - Virtual Foundation for Advancement of Science and Technology (Pakistan)
Not a member yet
1255 research outputs found
Sort by
Emphasis through Extra Particles: The Issue of Additional Elements in the Holy Qur’an in the Light of Tafsīr Abī al-Su‘ūd – The Case of the Redundant Particle Lā as a Model
Full text linkThis research explores the phenomenon of emphasis through additional particles (al-tawkīd bi-l-zawāʾid) in the Holy Qur’an, with particular attention to the particle lā, which is 70 often considered redundant, as interpreted in Tafsīr Abī al-Suʿūd. The study investigates whether these particles should be viewed as superfluous elements or as purposeful rhetorical instruments that enrich the eloquence and expressive power of the Qur’anic discourse. By drawing upon the exegetical perspectives of Abū al-Suʿūd in conjunction with classical sources on Qur’anic sciences, Arabic linguistics, and rhetoric, this work sheds light on the semantic and stylistic functions of lā in selected verses. The findings reveal that such particles are not mere insertions but serve as emphatic devices that strengthen meaning, maintain rhythmic balance, and enhance the aesthetic quality of the Qur’an’s language. Positioning Abū al-Suʿūd’s views within the wider exegetical and linguistic tradition, the paper highlights the critical role of linguistic and rhetorical analysis in uncovering the profound layers of Qur’anic expression and illustrates how these linguistic features reaffirm the Qur’an’s unique inimitability (iʿjāz)
From AI Ethics to AI Justice: A Comprehensive Framework for Equitable Governance in Education and Social Systems
Full text linkArtificial intelligence is embedded in educational systems and social structures, which creates ethical issues that cannot be resolved at the algorithm level only. This paper provides a critical review of general AI ethics and forms an AI justice angle based on relational equality. It systematizes the criticism into three spheres: (a) conceptual - misleading labels and narrow frames of understanding that conceal the infrastructural reach of AI; (b) substantive - ignored environmental, labour and distributional harms in AI life-cycle; and (c) procedural - advisory processes where the industry dominates and expertise is limited. Based on these diagnostics the paper proceeds with actionable reforms: use accurate language and situational analyses; consider non-technical options first, and then AI solutions; involve community and civil-society actors in deliberation; and instigate sound philosophical ethics into interdisciplinary teams. The contribution of the paper is a practical justice-driven roadmap that can be used by educators, policymakers and social scientists to assess AI implementation and direct the process toward more equitable and sustainable results
Linking Personality Traits to Vocational Proficiency: A Study of TVET Learners Using Advanced Feature Engineering
Full text linkThis study presents a comprehensive analysis of personality traits and learning skills among Technical and Vocational Education and Training (TVET) learners, leveraging advanced feature engineering and robust statistical methods. The primary objective is to investigate the influence of personality traits on learning outcomes and map these traits to specific skills across various TVET trades. Using a four-step methodology, we collected demographic data and Big Five Inventory (BFI) personality trait scores from 1,076 trainees across eleven Vocational Training Institutes (VTIs), specializing in eighteen distinct trades. Descriptive statistics, Spearman’s rank correlation analysis, and trade-specific BFI numerical scaling were employed to examine the relationship between personality traits and skill proficiency. Key findings reveal: (i) significant correlations between specific personality traits and trade proficiency, with notable variations across age and gender groups, and (ii) successful skill-to-trade mapping using BFI traits. These results highlight the potential of personality-aware approaches in vocational training to enhance occupational diversity and economic integration. By demonstrating the importance of aligning individual traits with skill development, this study contributes to the growing discourse on personalized vocational education. Future research should explore longitudinal studies to assess the long-term impact of personality traits on skill acquisition, career progression, and adaptability in vocational settings
Gamified Learning Application for Students with ADHD in Pakistan: A Learning-Based Experiment
Full text linkThe paper discusses how gamification can be used to enhance the learning performance of students with ADHD in Pakistan. The research promotes the importance of gamification as one of the tools that might be used to assist students who tend to experience challenges in the areas of concentration, motivation, and conventional learning settings. The authors created a prototype of a gamified learning application and went through with it extensively on a group of 146 students, aged 13-17 years. The app was created in a way that promoted participation, the rewarding of progress, and that kept the person engaged in the learning process.The research results revealed that the application was successful in enhancing the focus, interest, and learning results of the students. Students who applied gamified learning app showed high improvement as compared to the control group who did not apply the gamified learning app. The findings have indicated clearly that gamification is a viable strategy that can be used to enhance the learning outcomes of ADHD students.What is more important, the research noted the app was especially effective among the students who had troubles with staying concentrated and motivated in the traditional learning conditions when distraction could diminish their performance. The gamified learning application was also demonstrated to be quite effective among students suffering ADHD with poor basic mathematics and reading abilities and offered them an effective and supportive learning experience.In sum, it is possible to state that the given paper has confirmed that gamification is a useful educational resource in Pakistan. Gamified learning app will help ADHD students to have a better focus, get more motivated and have improved academic results, which makes the process of learning more meaningful and effective
Impact of Online Learning on Students’ Academic Performance: A Comparative Study of Online and Face-to-Face Learning
Full text linkThis study investigates the impact of online learning on students\u27 academic performance in comparison to traditional face-to-face learning, focusing on higher education institutions in Mardan, Khyber Pakhtunkhwa, Pakistan. The COVID-19 pandemic accelerated the shift toward virtual education, yet disparities in technological infrastructure, digital literacy, and instructional methods created challenges for students and educators alike. Employing a mixed-methods approach, the research collected data from 200 undergraduate students through structured questionnaires and interviews to evaluate key indicators such as academic achievement, engagement, motivation, and technological accessibility.Results revealed that face-to-face learners consistently outperformed their online counterparts in terms of GPA, participation, and comprehension. Online learning was significantly hindered by unreliable internet access, limited availability of devices, and insufficient teacher training in digital pedagogy. Furthermore, students reported difficulties in maintaining attention, managing self-directed learning, and engaging with instructors in online environments.The study concluded that while online learning can supplement education during emergencies, it lacks the structural and interpersonal benefits of conventional classroom settings, especially in under-resourced areas like Mardan. It was suggested that using a mix of learning methods, training for teachers and building better digital tools will boost the results of online education.This thesis supports educational policy talks by suggesting ways to close the gap in digital access and enhance student results using online approaches
Predictive Job Category Classification Using Deep Learning and Transformer Approaches
Full text linkIn the dynamic employment environment, job trials where the type of job is predicted based on the job description is very essential in minimizing the challenges in the job market. This paper discusses how machine learning and deep learning can be used to enhance job type prediction by solving ambiguity in job descriptions. With the help of supervised learning, we process both structured and unstructured data to facilitate and improve the accuracy of categorization and suggest more qualified candidates to be hired. Comparative analysis of machine learning algorithms, such as Logistic Regression, Naive Bayes, Support Vector Machines, Random Forest, and Gradient Boosting, and deep learning models, such as Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Recurrent Neural Networks (RNN), demonstrate that major progress has been made in precision, recall, and F1 scores of job matching. Random Forest has the best accuracy of 0.85, whereas GRU having a three-layer architecture has competitive results with an accuracy of 0.60, whereas BERT Large has the same accuracy as Random Forest of 0.85. These improvements are demonstrated on a heterogeneous dataset, enhancing the relevance of recommendations. Challenges such as ambiguous job descriptions and the rapidly changing labor market were identified as potential barriers. The study highlights the potential of machine learning to automate recruitment processes, reduce traditional CV screening costs, and improve job seeker experiences. This work pioneers adaptive job recommendation models with the potential to revolutionize global recruitment
Systematic Analysis of Search-Based Strategies for Combinatorial Test Suite Construction
Full text linkThis systematic literature review (SLR) investigates search-based strategies for generating combinatorial test suites using covering arrays (CAs) to efficiently test system interactions. Conducted following PRISMA guidelines, the review analyzes 91 primary studies published between 2003 and 2025, selected through a rigorous process from major academic databases. The identified strategies are categorized into five types: standard, mix, adaptive, hybrid, and hyper-heuristic, based on their underlying algorithmic approaches, including swarm intelligence, evolutionary algorithms, and hyper-heuristic techniques. Each strategy is examined in depth, evaluating its effectiveness in generating high-quality combinatorial test suites. The review also highlights challenges in applying these strategies to varying software testing scenarios. Based on the findings, it provides practical insights to enhance their application and effectiveness in real-world contexts. This work supports broader adoption of search-based testing to improve software quality and reduce defect rates
Enhanced Malicious Website Detection Using Machine and Deep Learning Techniques
Full text linkMalicious websites currently pose an escalating security threat that leads to malware contaminations and leaking of sensitive data. The customary detection methods generate numerous incorrect alarms while struggling to identify optimized security risks, so better detection techniques become essential. A dependable detection system for malicious websites is proposed, which utilizes AdaBoost and XGBoost algorithms along with Convolutional Neural Networks (CNN) by integrating machine learning techniques. The four-phased framework as suggested includes (1) Data Acquisition and Initial Analysis, which makes use of a dataset available on Kaggle to realize patterns; (2) Data Cleaning, Normalization, Segmentation, and Model Training processes, which clean, normalize, and segment the data in preparation for optimal model training; (3) Detection and Classification Evaluation, which checks against performance metrics such as precision, recall, F1-score, and accuracy; and (4) Comparative Study and Outcome Comparison checks against relevant literature and reports on CNN performing better compared to conventional techniques. The CNN model demonstrated successful accuracy performance at 96.58% during malicious URL identification processes with reduced occurrence of wrong positive detections. The improvement of this research lies in the introduction of the real-time contextual augmentation within the detecting architecture to enable the adaptive learned malicious URL detector with reduced instances of false alarm and greater flexibility towards new cyber vulnerability. Practical applications for malicious website detection can utilize advanced algorithms backed by the research findings, which provide guidance for future research. This research contributes by providing a robust comparative framework that establishes CNN as the most effective classifier for malicious website detection, reduces false alarms, and strengthens practical cybersecurity applications while offering guidance for future research. The research contributes by providing a robust comparative framework that establishes CNN as the most effective classifier for malicious website detection, reduces false alarms, and strengthens practical cybersecurity applications while offering guidance for future research
Interpretable Deep Learning for Brain Tumor Diagnosis: Occlusion Sensitivity-Driven Explainability in MRI Classification
Full text linkMagnetic resonance imaging (MRI) serves as a crucial diagnostic tool, particularly for brain tumors where early detection significantly improves patient prognosis. The growing use of deep learning in medical imaging has led to substantial progress, yet the opaque nature of these models creates barriers to clinical acceptance, especially for critical applications such as tumor diagnosis. Our research applies explainable AI (XAI) techniques to improve the transparency of CNN-based brain tumor detection using MRI data. Working with a dataset containing 7,022 images spanning four tumor categories, our model attains 80\% accuracy while employing occlusion sensitivity analysis to produce visual interpretations. These heatmaps identify the most influential regions for predictions, giving clinicians insight into the model\u27s decision process. This XAI integration enhances both understanding and accountability in healthcare AI systems, facilitating more reliable diagnostic tools.Precise early identification of brain tumors through MRI dramatically affects survival outcomes, though human interpretation remains time-consuming and variable. While CNNs show impressive classification results, their unclear reasoning limits clinical implementation. Our study introduces an XAI approach that pairs an accurate CNN classifier (80% on 7,024 multi-class scans) with occlusion analysis to create intuitive visual explanations. By methodically altering image areas and measuring prediction variations, we generate heatmaps that accurately pinpoint tumor-distinguishing features, matching radiological assessment. Comparative results demonstrate occlusion analysis\u27s superiority over gradient methods like Grad-CAM in spatial precision for tumor classification (meningioma, glioma, pituitary). This research progresses clinically useful AI by connecting model effectiveness with interpretability in brain tumor imaging
Assistive Living in IoT-Based Smart Home Systems for Aging in Place: A Review
Full text linkThe fast expansion of the elderly population has raised the need for creative solutions to help senior people live independently and securely at home. Internet of Things (IoT) based smart home systems rising as a promising solution to meet this need, provide health monitoring, fall detection, and daily activity recognition help among other features. Designed for aging in place, this paper offers a thorough review of IoT based smart home technologies, assessing their uses, advantages, and obstacles. Key technologies we find via a methodical literature review include wearable devices, environmental sensors, and voice activated assistants; we assess their efficacy in improving the quality of life for elderly people. The review points out important concerns - usability, data privacy, interoperability, and cost - that impede the general use of these systems. We also consider the consequences of these results for researchers and offer direction for further study to fill current research gaps. The possibility of IoT based smart home systems to help aging in place is underlined in this research, but the need for user-centered design, interoperable system, strong security precautions and inexpensive solutions to guarantee their accessibility and efficacy for aging populations is also stressed