| 題名 | Prediction Interval of Software Fault-Detection Time Using Back Propagation Algorithm. |
| 著者 | *Momotaz Begum, Tadashi Dohi (Hiroshima University) |
| キーワード | software reliability, fault prediction, prediction interval, Back propagation algorithm, delta method |
| アブストラクト | Software reliability is an important quality measure that quantifies the operational profile of computer systems. We derive the prediction interval of the software fault-detection time in sequential software testing, by applying the well-known back propagation algorithm and the delta method to construct the prediction intervals. In numerical experiments with two real software development project data sets, we evaluate the one-stage look ahead prediction interval of the fault-detection time with different network architecture, and compare them in terms of coverage rate and prediction interval width. Numerical results show that our proposed approach is rather robust across different software projects and has a better performance with respect to both goodness-of fit and step-predictability for software failure time. |
| 題名 | A Note on Mean Time to Security Failure of VM-Based Intrusion Tolerant Systems |
| 著者 | *Junjun Zheng, Hiroyuki Okamura, Tadashi Dohi (Hiroshima University) |
| キーワード | VM-based intrusion tolerant system, security, mean time to security failure, Markov analysis, Poisson process |
| アブストラクト | Nowadays, computer system faces the threat of security intrusions due to malicious attacks. It is important to develop the mechanism to tolerate intrusions. For example, we presented a generalized scheme to tolerate intrusions for virtual machine (VM)-based intrusion tolerant system [1](Zheng,2015). On the other hand, security is treated as a QoS (Quality of Service) attribute at par with other QoS attributes such as performance. This paper considers a security measure called ``mean time to security failure (MTTSF)'' for the VM-based intrusion tolerant system in [1]. |