VFAST - Virtual Foundation for Advancement of Science and Technology (Pakistan)
Not a member yet
1255 research outputs found
Sort by
Comparative Analysis of Exchange Rate Forecasting Techniques: Emphasis on Machine Learning Algorithms for Pakistan
Full text linkThe exchange rate is crucial because it can influence a country’s economy. It helps brokers and traders in operational decisions that help reduce risk and maximize profits. Many methods of forecasting currency exchange rates exist. The present study focused on different methodologies, including Box-Jenkins, Holt’s practice, artificial neural networks, Facebook Prophet Model, and Multilayer Perceptron (MLP) for predicting exchange rates. The performance of these techniques is evaluated based on small mean squared error, mean absolute error, and mean absolute percentage error. The results revealed that MLP outperformed all the models. It is a promising method to forecast the exchange rate of Pakistan because it gives a minor forecast error. In addition, the predicted values using MLP are very close to the actual values. The experimental results and time series plot revealed that the exchange rate of Pakistan will slightly increase in the upcoming months. It is concluded that the present study will help to determine the aggregate demand for domestic currency in the coming months. It is also helpful for the government and policymakers. However, understanding exchange rates is essential for anyone involved in international business and finance
Integrating COBIT 2019 with Zero Trust Architecture: A Strategic Approach to GRC in Cybersecurity
Full text linkThis paper is the first to show a shared security governance model that combines COBIT 2019 for strategic oversight and ZTA for tactical, real-time enforcement. This model addresses the poor performance of traditional GRC and perimeter security that cannot handle distributed clouds and sophisticated insider attacks. We posit that the objective of COBIT strong governance could be operationalized with ZTA’s “never trust, always verify” principles. The present study illustrates the integration by linking COBIT’s process domains (EDM, APO, BAI, DSS, MEA) to the core ZTA pillars (Identities, Devices, Networks, Applications, Data). This mapping provides confirmation that ZTA is a practical mechanism for control assurance that enables organizations to take a strong, risk-sensitive, and compliant posture within a changing digital environment
The Secure GPS Tracking Data for transportation in Distributed Environments
Full text linkEfficient and secure tracking of the public transport has become a critical issue in smart cities, especially in urban areas like Karachi which has high population. Current systems are usually characterized by slowness, unreliability in data transmission and susceptibility to security risks including services by unauthorized users and information alteration. In an attempt to solve these problems, this paper introduces a Secure Transporting Tracking Method one of the (SSTM) which is a new AI-based system of real-time GPS tracking of the public transport in the distributed environment. The SSTM incorporates a secure transmit of GPS data, AIs, location prediction, and efficient vehicle-passenger matching to make it more accurate, less time-consuming, and data integrity-insured. The model has been applied to and tested in several places all around Karachi, such as Gulberg, Nipa, Gulshan-e-Iqbal and Sachal. The outcomes of simulations prove that SSTM is faster than the current approaches, such as the traditional GPS and chatbot-enhanced ones, in processing, encryption/decryption, and tracking. The research brings in a safe, large scale, and smart transport tracking platform specific to smart cities with the possible use in real-time fleet management and secure passenger safety systems
Buddy Finder: A Privacy-Preserving Clustering Approach for Like-Minded Social Connections
Full text linkApplications that connect users with like-minded individuals while safeguarding their privacy are in high demand due to the quick growth of digital platforms for social interaction. Buddy Finder, a method for locating potential friends based on shared interests and physical proximity, is presented in this work. In contrast to conventional social networking or dating websites, the system prioritizes user privacy by displaying only a small amount of information during the first stages of interaction. Users are grouped based on their shared interests using clustering techniques, with K Means serving as the main algorithm. Participants in these groups may start conversations and gradually decide what more details to disclose. For use with a variety of data sets, the clustering code is effective, scalable, and adaptable. Additionally, the system is intended to facilitate meetings inside the local community as well as meetings that take place locally. In addition to outlining the overall architecture design, algorithmic design, and implementation difficulties, this study provides recommendations for future advancements in personalization and privacy recommendation systems
A Review on Cardiovascular Diseases Risk Prediction Approches Based on Machine Learning and Evolutionary Algorithms
Full text linkAcross the world, 17 million people die from heart disease each year. Heart-related diseases were the main cause for about 19\% percent of deaths in Pakistan in 2016, the same has since now risen to 29%. As per most recent WHO statistics regarding prevalence of heart attacks in Pakistan, approximately more than two hundred thousand (200000) persons died in Pakistan in 2020 due to coronary heart disease, making up 16.49 percent of all fatalities. With a death rate of 193.56 per 100,000 inhabitants, Pakistan is ranked at number 30 in the world. Rising death rate due to heart disease can be minimized through detection at early stage. Different data mining approaches have made early detection of cardiac disease possible. Certain datasets are being used to retrieve useful information. Several machine learning techniques / models have been proved to be the most effective, accurate and profitable to detect cardiovascular disease at an early stage. However, the approach of machine learning and Genetic Algorithm (GA) with feature selection may aid in lowering the computational complexity of GA and increasing the effectiveness of its search for ideal solutions. Hence, there is dire need to apply such hybrid approach to get much more effective and accurate results. The goal of this survey paper is to review different papers related to CVD prediction at early stage by applying hybrid approach of ML Techniques with Genetic Algorithm. Moreover, the results obtained by the authors in reviewed papers are also examined. In the end, this survey will showcase the importance of the hybrid approach in improving accuracy of ML results
DermInsight: A diagnostic System for Human Skin Diseases utilizing Deep Learning
Full text linkSkin diseases present significant challenges in healthcare due to their diverse diagnosis and treatment. They ranked among the top causes of disability because most people ignore their early symptoms due to costly and time-consuming diagnostic methods, which worsen the skin condition with time. This research-based project is a contribution to the timely and cost-effective identification of seven prevailing skin diseases. We propose a novel deep learning-based system capable of classifying distinct skin diseases including melanoma, melanocytic nevi, actinic keratosis, benign keratosis, basal cell carcinoma, dermatofibroma, and vascular lesions. By leveraging a comprehensive dataset HAM10000, our system achieved an impressive accuracy rate of 98.14% in accurately identifying and categorizing these skin diseases. We employ transfer learning and fine-tuned three advance deep learning models MobileNetV1, MobileNetV2, and Xception, and evaluate their performance in the classification of seven human skin diseases. Remarkably, MobileNetV1 emerged as the top-performing model, surpassing the capabilities of the other models and existing state-of-the-art methods. In addition, the proposed model is deployed in an android app named “DermInsight” for the use of dermatologists. The dermatologist uploads the image of the skin lesions of a patient and the app will predict the disease within 1 to 2 seconds
A Behaviorally-Driven Software Architecture for E-Government Adoption in Pakistan
Full text linkE-Government services were launched in Pakistan in October 2002, marking a significant step toward digital governance. While policy initiatives and infrastructure development have continued, the successful adoption of these services remains constrained by several socio-technical and behavioral challenges. This study presents a user-driven software framework for e-government adoption in Pakistan, aimed at bridging the gap between user behavior and system design. By integrating behavioral insights into system architecture, the framework emphasizes the importance of user acceptance, trust, and perceived ease of use in shaping e-government adoption. A theoretical model was developed based on an enhanced combination of three established models—Technology Acceptance Model (TAM), Unified Theory of Acceptance and Use of Technology (UTAUT), and the Information System Success Model—to explore the critical success factors influencing e-government implementation. The model was validated through empirical data collected from computer professionals working in Pakistan’s government sector, particularly in education and health departments. A structured questionnaire comprising 42 items across 10 constructs was distributed to 650 respondents, yielding 508 valid responses. The results revealed strong support for all proposed hypotheses and confirmed the role of behavioral intention as a key driver of adoption. This study not only confirms the relevance of behavioral factors in e-government adoption but also proposes a modular and scalable software framework that can inform the design of more user-centric, secure, and behaviorally-aligned e-government platforms. The findings offer practical implications for policy makers, developers, and system designers seeking to improve e-government engagement and digital service delivery in developing countries like Pakistan
The Impact of AI Tools (Like ChatGPT) on Student Learning Outcomes and Academic Integrity: A Mixed-Methods Study
Full text linkThe fast pace of the introduction of artificial intelligence (AI) tools (including ChatGPT) into the educational process has already raised a storm of controversy about the impact they have on student achievement and academic integrity. This combined research evaluates the impact of writing assistants powered by AI on academic achievements and critical thinking abilities and moral reflections of students as well as within the educational landscape. Surveys of 300 university students were used to collect quantitative data on their habits regarding the use of AI tools, perceived benefits and challenges of using them. It yielded qualitative data through a qualitative survey concerning 15 educators through in-depth interviews to discuss the responses of institutions, policy gaps and pedagogical issues. The preliminary findings suggest that although the AI instruments can increase the efficiency of research and the drafting process, they are also associated with the issue of excessive dependence, lack of original thinking, and the risk of plagiarism. There was a notable correlation between a high frequency of AI use and positive short-term educational achievements, yet long-lasting components of the cognitive effects are not certain in any way. Teachers had rather ambivalent opinions, with some supporting the introduction of AI literacy at education institutions, and others demanding enforcement and regulation. The paper underlines the importance of a balanced AI implementation- stimulating progress and protecting the campus integrity. Suggestions such as the creation of institutional rules, AI-detective devices, and critical thinking modules should be created in order to reduce the misuse. In terms of the continued debate about AI in education, this study adds to the empirical literature to be used by policy and practices in the age of digital learning
Vocal Sentiments: Transformer Based Speech Emotion Recognition
Full text linkSpeech Emotion Recognition (SER) plays a crucial role in Human–Computer Interaction (HCI) by enabling systems to interpret and respond to human emotions through speech analysis. This paper presents a Transformer-based SER framework that leverages the Wav2Vec2 model for self-supervised representation learning. Unlike conventional approaches relying on handcrafted acoustic features or shallow learning, our approach employs transfer learning to extract high-level contextual embeddings from raw audio. We integrate two benchmark datasets, RAVDESS and TESS, to improve generalization across diverse speakers and emotions, and further analyze system robustness by introducing varying levels of environmental noise. The proposed model achieves an accuracy of 79.01%, with balanced precision, recall, and F1-scores, demonstrating competitive performance compared with recent state-of-the-art SER models. The main contributions of this work are threefold: (i) a novel evaluation of Wav2Vec2 embeddings on combined RAVDESS–TESS data, (ii) a systematic assessment of noise robustness in Transformer-based SER, and (iii) a comprehensive benchmark that highlights the strengths and limitations of transfer learning in practical emotion recognition scenarios. These findings suggest broad applicability in voice assistants, call-center analytics, and mental health monitoring, while future extensions may incorporate multimodal data and advanced fine-tuning strategies to further enhance performance
Role of IoT in Disaster Monitoring and Response: A Comprehensive Review
Full text linkIn an age where natural and man-made disasters are increasing in frequency and intensity, the impact of technology is expanding in disaster management. One of these transformative technologies is the Internet of Things (IoT) in disaster management. IoT-based disaster management systems involve monitoring environmental conditions, such as temperature, humidity, seismic activity, rainfall, water levels, or gas concentrations, by utilizing sensor nodes deployed in the field to autonomously record relevant data. This recorded data is then transferred to a central sink node or base station, performing as a gateway node for further communication, data collection, and smart decision-making. This review article comprehensively explores the critical aspects of the Internet of Things (IoT) in off-site disaster monitoring and management. It offers a broad overview of IoT-enabled disaster management applicable across various disaster types, including earthquakes, floods, forest fires, and nuclear hazards, consisting of sensor networks, supported by wireless communication, cloud computing, and advanced data analytics, that provide real-time understanding, foresight, and automated responses across all four phases of disaster management, including mitigation, preparedness, response, and recovery. It also highlights the importance of integrating IoT with emergent technologies such as artificial intelligence (AI), edge computing, big data, and 5G, which can significantly increase the responsiveness, efficiency, and resilience of disaster management systems