Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
3 results
Search Results
Item Machine Learning Approach for Testing the Efficiency of Software Reliability Estimators of Weibull Class Models(Springer Science and Business Media Deutschland GmbH, 2025) Murulidhar, N.N.; Tantri, B.Usage of software in every field has resulted in the concern over its quality and durability. In this regard, there is a need to have a systematic way of assessing the reliability of the software. One such assessment is the estimation of software reliability. Numerous works have been done in estimating the reliability of the software by making use of software failure times. Most of the software failure times follow Weibull distribution. Herein, Weibull models are considered. Two well-known estimators, viz, the Maximum Likelihood Estimator and the Minimum Variance Unbiased Estimator have been obtained and combined to get Improved Estimator, which satisfies maximum number of statistical properties of a good estimator. In addition, the comparison of the three estimators is carried out by means of coefficient of variation, which considers both the mean and the standard deviation. The comparison is further enhanced by applying statistical tests to these estimators. Machine learning, being the most widely used technique in recent times, herein it is intended to use the R programming language, which is considered as a powerful machine learning language, to carry out statistical tests pertaining to these estimators. Few datasets have been considered and the estimates have been tested for comparison using Modified signed-likelihood ratio test for equality of coefficients of variations. The output results have been analyzed to test the significance of the differences between the coefficients of variation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.Item Novel Software Reliability Estimate for Exponential Class Models(International Society of Science and Applied Technologies, 2022) Murulidhar, N.N.; Tantri, B.R.Increasing usage of software in every domain has raised concern over its quality and durability. Many indicators for measuring the quality and durability of the software exist. One such indicator is the software reliability, which is a measure of the life time of the software. Estimation of software reliability enables the users of the software to decide whether or not to accept the software. Knowing the probability distribution of the failure times of the software, the reliability of the software can be estimated. Herein, software reliability models having exponential failure times have been considered. The reliability has been estimated by considering the methods of Maximum Likelihood Estimation (MLE) and Minimum Variance Unbiased Estimation (MVUE). The two estimators are combined to obtain the Improved Estimator (IM). Few data sets have been considered and the estimates have been obtained using the said three methods. The three estimators are then compared using the coefficient of variation. It is observed that the Improved Estimator possesses the least value of coefficient of variation, thus indicating that the Improved Estimator is better as compared to the other two estimators and hence provides more accurate estimate of reliability. © 2022 International Society of Science and Applied TechnologiesItem Improved Estimator of Software Reliability for Weibull Class Models(International Society of Science and Applied Technologies, 2023) Murulidhar, N.N.; Tantri, B.R.Increase in the usage of software in every field has resulted in having concern over its quality and durability. Research in this area is still of importance and many researchers are still working towards the improvement in the reliability of the software products. Measures of quality in terms of reliability are vast and obtaining the estimate of reliability would provide more insight into the durability and hence in assessing the performance of the software. Software reliability models are widely used in this estimation process. Most of the failure data models fall into Weibull class models, in which, the failures times are assumed to be distributed as Weibull. Herein, such Weibull class software reliability models are considered. It is intended to combine two well-known estimators, viz, the Maximum Likelihood Estimator and the Minimum Variance Unbiased Estimator. Both estimators have their own pros and cons, in terms of the properties satisfied by them. Herein, it is intended to preserve the statistical properties satisfied by both the estimators by combining them to get an Improved Estimator, which satisfies maximum number of statistical properties of a good estimator. In addition, the comparison of the three estimators is carried out by means of coefficient of variation, which considers both the mean and the standard deviation. The comparison is further enhanced by considering the quartile coefficient of dispersion of the three estimators. Some bench mark failure data are considered to establish the efficiency of the improved estimator. © RQD 2023. All rights reserved.All right reserved.