VFAST - Virtual Foundation for Advancement of Science and Technology (Pakistan)
Not a member yet
1255 research outputs found
Sort by
Improved Population Proportion Estimators via Exponential Term and Twain Auxiliary Attributes in Simple Random Sampling
Full text linkThis article presents a innovative group of ratio regression estimators utilizing exponent to improve the accuracy of estimating finite population proportions .These estimators enhance precision by incorporating twain ancillary characters under a simple random scheme (SRS). First order approximations have been employed to establish the theoretical relationships for the bias and Mean square Error (MSE) of the proposed estimators. To evaluate their performance, a real-world data set is analyzed using MSE and percent relative efficiency (PRE) as key evaluation metrics. The results indicate that the recommended estimators exhibit enhanced efficiency relative to several existing estimators discussed in the literature. Empirical analysis supports the theoretical findings, providing strong evidence of the improved statistical performance of the recommended class of estimators over conventional alternatives
Deep Learning-Based Image Segmentation Techniques for Bone Fractures Using X-Ray Images: A Systematic Review: Deep Learning-Based Image Segmentation Techniques for Bone Fractures Using X-Ray Images: A Systematic Review
No full textHuman bone fractures are common musculoskeletal disorders. The primary cause of fractures is often accidents or external pressure applied to the body, which can result in significant fractures. Medical image processing plays a crucial role in the segmentation and analysis of human bone fractures using X-ray images, assisting physicians in determining appropriate treatments. The use of artificial intelligence (AI) techniques, such as machine learning, deep learning (DL), and transfer learning, has garnered significant interest for medical diagnosis from X-ray image reports. The primary objective of this paper is to explore various deep learning-based methods for analyzing human bones using X-ray images. These methods include the evaluation of U-Net, ViT, TransUnet, Swin-Unet, and Swin-Unet++, with U-Net and SegNet being utilized for comparative analysis. The findings and discussion indicate that U-Net and ViT are among the most promising models for the MURA dataset, achieving high accuracy. A comparison chart is provided in the paper to highlight various fracture segmentation methods, dataset sizes, and evaluation metrics
A Systematic Approach to Probabilistic Modeling for Retrieving Sindhi Text Documents
Full text linkInformation Retrieval (IR) systems retrieve the relevant information from the document based on the user queries. However, due to the paucity of linguistic resources, IR systems for the regional languages of Pakistan failed to achieve the required accuracy. A new systematic methodology is presented in this research article to retrieve the text form the document written in Sindhi language using Probabilistic modeling approach. This approach is considered as most efficient accessible methodology in most of the probabilistic models. Following the data collection, total 108 newspapers and 29 books were categorized into text documents against different topics. Collectively, 68765 documents were generated in all. For test cases, twenty queries applied to each case categorized into three classes i.e. One-Word query, Two-Word query and Three-Word query. Total 136 documents retrieved which provide 0.79 average precision cumulatively. However, further documents need to be added into the developed database in order to improve the performance of the IR system
Investigating Effort Estimation Techniques for Mobile Applications: An Efficient Approach
Full text linkEstimating the effort of mobile applications is essential because many of the applications are now working on mobile platforms. A need exists to understand the difference between Effort Estimation for mobile applications and other computer applications. The last decade has seen a revolution in the use of mobile applications, which has caused in an exponential increase in the total number of mobile phone users worldwide. The first objective of this work is related to the software industry, and that is to identify which techniques are used for calculating the effort of mobile applications. The first objective also dwells into the identification of the accuracy that was achieved by using those techniques. The second objective is to propose an efficient approach for the effort estimation of mobile applications. A 5+1 methodology is suggested which should be accommodated when proposing a model for the effort estimation of mobile applications. The proposed methodology is validated through intensive investigation of the literature and it is believed that if this 5+1 methodology is adopted, the proposed model will surely bring excellent results in terms of accuracy of the predicted effort that the proposed model will attain. A small case study is also mentioned as a starting point for the validation of the 5+1 proposed methodology and it shows how the methodology can be utilized for the effort estimation of a simple mobile application
A Review of Skin Disease Detection Using Deep Learning
Full text linkAmid increasing concerns about skin diseases exacerbated by climate change or lifestyle, some diseases are undiagnosed or misdiagnosed due to limited healthcare facilities. The worldwide health burden emphasizes the need for innovative diagnostics. This study explores the evolutionary role of deep learning in skin disease detection, providing the most advanced and effective research approaches, model achievements, and dataset usage exclusively. The review adapts data from 30 research papers and many datasets to address imbalanced class and various efficiency factors. The developments in CNN models like MobileNet or EfficientNet, have strengthened computational potential, while hybrid models have accommodated local and global features. Furthermore, Explainable AI (EXI) and augmented datasets have overcome the challenges including noisy, biased datasets and the less interpretable AI models. This study declares the innovative capacity of deep learning in dermatological analysis, highlighting its scalability and performance. Future research is required to consider dataset diversity, interpretability, and incorporating medical metadata to enhance model performances
The Deepening Gaza Crisis: Humanitarian Struggles and International Response to Post-October 7, 2023
Full text linkThe crisis between Israel and Palestine has been so long and has brought immense suffering. This paper focuses on the humanitarian crises, and civilians of Gaza from October 7, 2023, till mid-September 2024. Firstly, a brief history of the conflict with key events that led to the attack of October 7, 2023. It also covered the mass attack by Hamas on Israel that claimed the death of more than 1,200 Israelis. In response, Israel launched "Operation Iron Sword," which threw Gaza into a state of war. This paper highlighted the incident in which the collapse of hospitals, infrastructure, and schools and the shortage of food, water, and medical supplies has been covered. This is clearly the violation of human rights, particularly the unjust detention and torture of civilians, mainly women and children. The paper also provokes the comparison of the condition between hostages taken by Hamas and the arrest of civilians in Gaza by Israel. Also, it observed the international response has been in two faces: the majority of countries condemn the actions of Israel while the rest support the actions in multiple ways. Lastly, the paper presents the most possible ways to resolve the conflict between nations, such as two-state and one-state solutions and alternate ideas like joint peace cities. The study emphasizes the urgent need to bring peace and end this ongoing havoc and to focus on how both parties could live coherently because bloodshed, bombings, and massacres had never been a solution
Reducing the Risk of Cyber Attack in SDN Network by using Blockchain
Full text linkIn the contemporary digital landscape, Software-Defined Networking (SDN) has emerged as a transformative approach that decouples network control from hardware, enabling greater flexibility and centralized management. However, this innovation has also introduced new vulnerabilities, making SDN networks susceptible to cyber-attacks. To reduce the risk of cyber-attack blockchain can play vital role. In paper we propose a novel framework that leverages blockchain technology to enhance the security and resilience of SDN environments against potential threats. By integrating blockchain’s decentralized and immutable characteristics, our approach facilitates secure data transactions, enhances network visibility, and fosters trust among network participants. We present a comprehensive analysis of the vulnerabilities inherent in traditional SDN architectures and outline method, how blockchain can mitigate these risks through secure authentication, data integrity, and enhanced access controls.Furthermore, we conduct a series of experiments to evaluate the performance impact and security benefits of our proposed solution. The results demonstrate a significant reduction in the likelihood of cyber-attacks, showcasing the viability of blockchain as a potent tool for safeguarding SDN networks. Our findings underscore the importance of interdisciplinary approaches in addressing the evolving challenges of network security, paving the way for more resilient and secure SDN infrastructures in the future
Deconstructing Class Struggle and Gender Oppression: A Post-Structuralist Analysis of Jamal Abro\u27s Pirani and Naseem Kharal\u27s Thirty-Four Gates through Barthes\u27 Five Codes
No full textThis study applies post-structuralist theory to analyze Jamal Abro’s Pirani and Naseem Kharal’s Thirty-Four Gates, focusing on how both stories portray class struggle and gender oppression within Sindhi society. Using Roland Barthes\u27 five codes—hermeneutic, proairetic, semantic, symbolic, and cultural—the research uncovers multiple layers of meaning, interpreting the symbols, tensions, and ambiguities that reflect the socio-political realities of a male-dominated, feudal, and capitalist system. By employing a qualitative textual analysis, this research explores how Barthes\u27 codes allow for the deconstruction of dominant narratives, revealing the complexities of class exploitation and cultural repression. The findings suggest that both authors critically condemn anti-women traditions and expose the systematic subjugation of women, who are rendered voiceless in a patriarchal society. This work contributes to a deeper understanding of Sindhi literature, translated into English and published by Oxford University Press, by engaging with the socio-political and cultural discourses embedded in the texts
A Review of Classification Approaches in Educational Data Mining for Predicting Student Performance
Full text linkWith the rapid increase in student data, and the growing interest in finding insights into student learning patterns,Educational Data Mining (EDM) methods are increasingly being used by educational institutes. Classification, a popular EDM method, enables the in-depth, efficient, and thorough analysis of student data while providing insights that directly assist in understanding student learning patterns and identifying elements that influence academic success. This review seeks to identify common trends and assess the effectiveness of four popularly explored classification approaches for predicting student performance. To assure the selection of research that specifically addresses the use of classification approaches for predicting student academic achievement, this review follows a systematic approach. A quality evaluation step was also included to help ensure that only reliable and credible studies were included in the review. According to the review findings of thirty two studies, most researchers used assessment results, academic performance index, and demographics to predict student performance. Decision Trees and Probabilistic classifiers were found to be the most popular and commonly used classification approaches for predicting student performance. The review also focuses on the challenges often faced while undertaking classification tasks in EDM and outlines future research directions in the context of analyzing student data
MLOps critical success factors - A systematic literature review
Full text linkMLOps encompasses a collection of practices integrating machine learning into operational activities, a recent addition to the diverse array of machine learning process models. The need to tightly integrate machine learning with information systems operations to ensure organizational performance led to the development of this approach. Therefore, MLOps methodologies are useful for businesses that want to make their ML operations and procedures more efficient. The purpose of this study is to summarize the many critical success factors that have been identified in studies focusing on MLOps initiatives. The paper shows how these CSFs affect MLOps performance and what factors drive this influence. We picked primary papers for analysis after conducting searches in three major publishing databases. We narrowed the field down to 58 unique CSFs, which were then classified according to three dimensions: technical, organizational, social and cultural. These CSFs affect and drive performance in MLOps, based on the results of the literature review. Researchers and industrial experts may enhance their understanding of CSFs and get insights into tackling MLOps difficulties inside organizations. The paper, notably, emphasizes several prospective research directions linked to CSFs