To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Empirical Role Rule Classification Model for Software Fault Forecast with Vector Machine Analysis |
|
Author |
Maaz Rasheed Malik, Liu Yining and Salahuddin Shaikh |
| Citation |
Vol. 19 No. 9 pp. 195-201 |
|
Abstract |
Our research aims to be analyses the software fault forecast with the help of machine learning and data mining tools. The analysis depends upon defected and non-defected datasets models. The datasets model we have used here are NASA datasets models. Our research proposed methodology is rule classification classifier with the help of vector machine. We have illustrated results in tp-rate, f- measure, area under curve (ROC) and correctly classified instances. Basically, these are measure efficiency unit which are used for measuring the accuracy and improvement of software fault forecast we have used here for analysis the proposed methodology vector machine with rule classification classifiers and without using of vector machine analysis. We observed that M5rule classifier is worst classifier in all over rule classification because it decreased his efficiency in all scenario case during the use of vector machine. But without using proposed solution methodology we can use it for analysis and can compare their results with other classifiers. ONER and PART classifiers are very good in all scenario cases because they have enhanced the efficiency and also improved the correctly classified instance c.c.i % ratio. |
|
Keywords |
Classifiers, Software Fault Forecast, Decision Table, Support Vector Machine, Rule Classification, Machine Learning, Data Mining. |
|
URL |
