Data Detection in Wireless Sensor Network Based on Convex Hull and Naïve Bayes Algorithm

Ramirez-Asis E.H.; Zapata M.A.S.; Sivakumaran A.R.; Phasinam K.; Chaturvedi A.; Regin R. · EAI/Springer Innovations in Communication and Computing (2023) · DOI: 10.1007/978-3-031-15542-0_3

Resumen

Wireless sensor networks (WSNs) have been affected by data due to their placement in random and risky atmospheres. Sensitive data in computer systems are increasing drastically and, thus, there is an utmost need to provide efficient cybersecurity. While detecting security bugs, software engineers discuss these bugs privately and they are not made public until security patches are available. This leads to many failures such as communication failure and hardware or software failure. This work aims to assist software developers in classifying bug reports in a better way by identifying security vulnerabilities as security bugs reports (SBRs) through the tuning of learners and data preprocessors. Practically, machine learning (ML) techniques are used to detect intrusions based on data and to learn by what means secure and nonsecure bugs can be differentiated. This work proposes a rudimentary classification model for bug prediction by involving Adaptive Ensemble Learning with Hyper Optimization (AEL-HO) technique. Classifier performance is analyzed based on the F1-score, detection accuracy (DA), Matthew’s correlation coefficients (MCC), and true positive rate (TPR) parameters. Comparisons are made among different already-existing classifiers.

Instituciones: Universidad Nacional Santiago Antúnez de Mayolo