VFAST - Virtual Foundation for Advancement of Science and Technology (Pakistan)
Not a member yet
1255 research outputs found
Sort by
Adaptive Bug Localization Framework for Precision-Driven Bug Localization in Software Engineering
Get PDFSoftware development always looks for automated methods to improve productivity and accuracy in issue detection. The paper conducts a comparative examination of several machine-learning techniques to tackle the bug localization difficulty. Our study compared the performance of Logistic Regression (LR), Random Forest Classifier (RFC), Support Vector Machine (SVM), Gradient Boosting Classifier (GBC), and Adaptive Bug Localization System (ABLS) on five dataset versions. The results demonstrate the superior performance of ensemble learning methods. The ABLS model regularly beats other models regarding F1 score, accuracy, and recall, indicating its strong potential for precise problem localization. The study highlights the necessity of continuously adapting models to tackle idea drift in dynamic datasets. Our research suggests a path for future endeavours involving improving feature engineering and integrating real-time online learning to sustain high performance in bug localization activities
Leveraging Machine Learning Models for Customer Churn Prediction in Telecommunications: Insights and Implications
Get PDFIn the world of telecommunications businesses, customer turnover poses a significant hurdle that can impact profits and weaken customer loyalty over time. Our solution to this challenge involves a method using Machine Learning (ML) tools to predict churn, with precision. We work with a set of 7In our research study we examined how well three different machine learning models performed. Random Forest (RF) Cat Boost (CB) and K nearest neighbors (KNN). Out of these models tested the Random Forest model stood out for its performance achieving 99 percent accuracy and precision along with an 88 percent recall rate and a 99 percent F1 score; additionally, it achieved an AUC of 0.99. These results clearly demonstrate the Random Forest model\u27s ability, in identifying customers who are likely to churn. The findings of this study hold importance for telecommunications companies as they are equipped with a valuable resource to proactively tackle customer turnover issues and customize solutions to retain key clients while boosting overall customer happiness levels in an increasingly competitive market landscape where keeping customers is crucial for business success our research provides a data supported roadmap for continual expansion and staying ahead in the telecom industry spotlighted in this abstract is the critical relevance of churn prediction for telecom firms underscored by the tangible advantages of leveraging the Random Forest model for predicting customer churn. By utilizing this advanced technology, telecom companies can proactively identify at-risk customers and take targeted measures to prevent them from leaving. This not only helps to retain key clients but also improves overall customer satisfaction. In a constantly evolving market, having access to predictive analytics can give companies a significant edge and ensure long-term success in the industry
A Novel Hesitant Cubical Dombi Fuzzy Aggregation Operators for Selecting Green Supplier Chain Managements
Get PDFA hesitant fuzzy (HF) set enhances the concept of fuzzy sets by addressing disagreements among decision-makers about the membership degree of an element. Similarly, the Cubical Fuzzy Set (CFS) is useful for managing uncertainty in decision-making problems. However, existing methods often lack integration of hesitation and cubical uncertainty, and there is limited exploration of their combined effects on aggregation processes. In this paper, we introduce the Hesitant Cubical Fuzzy Set (HCFS), which integrates the principles of HF sets and CFS to address these limitations. We define several set-theoretical operations for HCFSs and develop Dombi operations for them. Furthermore, we present a range of aggregation operators based on Dombi operations, including the Hesitant Cubical Dombi Fuzzy Weighted Arithmetic Averaging (HCDFWAA) Operator, the Hesitant Cubical Dombi Fuzzy Weighted Geometric Averaging (HCDFWGA) Operator, the Hesitant Cubical Dombi Fuzzy Ordered Weighted Arithmetic Averaging (HCDFOWAA) Operator, and the Hesitant Cubical Dombi Fuzzy Ordered Weighted Geometric Averaging (HCDFOWGA) Operator, and examine their properties. Additionally, we propose a multi-criteria group decision-making method and algorithm within the Hesitant Cubical Fuzzy framework. To address gaps in practical application, we provide an example of the selection of green suppliers in supply chain management. We also perform a comparative analysis with existing operators to highlight the advantages and effectiveness of our approach, emphasizing how the integration of hesitation and cubical uncertainty can enhance decision-making processes
FlightForecast: A Comparative Analysis of Stack LSTM and Vanilla LSTM Models for Flight Prediction
Get PDFThe Coronavirus was first reported in China in the city of Wuhan in December 2019, after a couple of months, it was widespread around the world. The whole world was in a state of lockdown. This hazardous disease affects the normal daily life of every individual and the tourism industry, especially the airline business was at a greater loss. Considering the airline business, this study contains data on commercial flights from 2019 to 2020. The conducted research analyzed the rise and fall of different flights in the lockdown period. The research is based on the variants of Long Short-Term Memory (LSTM) such as standard Recurrent Neural Network (RNN) and stack LSTM. The comparative research shows that the prediction of the stack LSTM model is better than the standard RNN keeping view of taking a considerable amount of time to train
Enhancing IoT Security through Machine Learning-Driven Anomaly Detection
Get PDFThis is study emphasizes the growing cybersecurity situations arising from the increasing use of Internet of Things (IoT) devices. Paying the main attention to the development of IoT security, the work here deploys the machine learning-based anomaly detection and adaptive defense mechanisms as proactive methods to counteract existing plus future cyber threat sources. The visual serves to expound the rapid development of the Internet of Things, and it also highlights the importance of infrastructures with robust safety features to secure the connected devices. IoT security statement brings out the hidden threat and vulnerabilities of the IoT, in this context advanced security measures are for the rescue. The objectives concentrate on improving security of IoT via machine learning detection of anomalies, and bring introduction of defense mechanisms that are adaptive. We specify the data sources, preprocessing tasks, and Random Forest, Decision Tree, SVM, and Gradient Boosting algorithms selected for anomaly detection in the methodology section. The abnormity negotiation function and the self-adaptive defense procedures are combined in order to strengthen the information technology ecosystems which are capable of dynamic simplification. The results and discussion part hotelates the effectiveness of machine learning models selected, and indicates about accuracy, precision, and recall metrics. To state in the most significant matter, Gradient Boosting brings the greater precision of 89.34%. Table 3 below indicates the various models\u27 effectiveness. It is proven that Gradient Boosting is the most powerful model among all. The discourse unfolds with account of the results, acknowledgment of the limitations, and discussion crucial obstacles encountered in the realization of the research. The conclusion reaffirms the importance of machine learning in IoT security implementation, thus building a robust system that can evolve to fight the ever-emerging cyber-attacks, keeping up with the progressive direction for securing IoT through the connected world
The Integration and Management of Clean Energy In Microgrid using Blockchain
No full textThe environment pollution is one of the biggest issues in the world and with the passage of time this issue becoming more severe. The researchers are focusing on this issue from many years but there is still no control and it’s creating bad impacts on weather and health. The impact of issue can be seen in form of changes occurs in global temperature, weather and health. There are many factors that contribute to raise pollution such as CO2, and main sources are smog, smoke and heat generated with energy resources. The pollution must be control for clean and healthy environment. The energy generate from the coal, diesel and by burning fuel must be stop and clean energy resources must be used such as hydroelectricity, solar and wind energy in future. In this research paper, the main issues related to energy and its environmental impact and solution will be proposed. The Microgrid is new approach and reliable and decentralized control system, which can address many technical and non-technical issues related to present power system by using the blockchain. A Simulink model of microgrid will be used for analyzing the result graphically.
Enhancing Efficiency and Safety with YOLOv5-Powered Robotic Arms for Waste Classification
Get PDFThe world is experiencing a transformation shift from manual labor to digital solutions, making work simpler and more efficient while enhancing the quality of life globally. A prime example of this shift is the Object Picking Robotic Arm (OPRA). Designed to operate with minimal human intervention, the OPRA reduces the risk of physical injuries among workers by replacing human labor with robotic precision. This technology finds applications in both industrial and domestic settings, including the automotive industry, metalworking, chemical processing, and various pick-and-place tasks. In this research, we develop a robotic system for automated waste picking and sorting. This system utilizes the YOLOv5 object detection algorithm to achieve high accuracy (95\%) and precision (90\%) in classifying five common waste categories: cardboard, metal, paper, plastic, and trash
Robust Hybrid Texture Descriptor (HTD) and a parallel score based fusion for face verification and liveness detection system
Get PDFCurrently, most of the biometric recognition systems are based on face verification are susceptible to the spoof attacks. Video replays, printed photographs and 3D mask attacks provoke false acceptance lest some necessary counter-measures are employed. We focus on still face spoof attacks considered as one of the most easily generated attacks and challenging for modern face verification systems. To detect face spoofing, most of existing countermeasures focus on designing discriminative features to analyze the textural properties of facial skin. To improve the texture discriminating properties and get advantages from other texture descriptor, in this paper, a novel Hybrid Texture Descriptor (HTD) is proposed and models the joint performance of face verification and liveness detection by score fusion method, for which the multi-modeling is well recognized. Numbers of experiments are carried out on UPM face spoof and one public domain Replay attack databases as standalone countermeasure in addition to the consolidation with verification system by means of fusion of score. The potential of proposed method demonstrate the outperformance throughout the experiments
Solution of nonlinear models in engineering using a new sixteenth order scheme and their basin of attraction
Get PDFThe utilization of computers in various engineering and mathematical fields has become increasingly important in recent times. In computational and applied mathematics, computer programs can be used to efficiently process complex data in mere seconds. By utilizing different computer tools and programming languages, computers can manipulate various iterative algorithms to solve nonlinear problems. The convergence rate and computational costs per-iteration are essential factors that determine the effectiveness of an iterative algorithm. This article introduces a new and highly efficient iterative approach of sixteenth order. The authors propose a four-step iterative optimal derivative-free method for solving non-linear equations that arise in engineering application problems. The proposed method achieves an accuracy of order sixteen with only five functional evaluations, and the efficiency index is 1.741. To demonstrate the performance of the method, the authors present several numerical examples, application problems, and a study of dynamics, comparing the method with other available methods of the same order in the literature. Numerical results and dynamics are obtained using computer tools
Stability and bifurcation analysis of a discrete Leslie predator-prey model with fear effect
Get PDFThis study examines a predator-prey model that includes the impact of fear and a square-root functional responseto represent herd behavior in the prey population. Our investigation aims to investigate the existence and stabilityof fixed points in this model. Through conducting an extensive analysis, we have uncovered valuable observations onthe model\u27s behavior, namely recognizing the occurrence of period-doubling and Neimark-Sacker bifurcations.These findings provide an understanding of the intricate dynamics that govern predator-prey interactions in the presence of fear and herd behavior. We provide numerical examples to support our conclusions