Please use this identifier to cite or link to this item:
|Title:||Development of Software Fault Prediction Models using Combined Metrics Approach|
|University:||Guru Gobind Singh Indraprastha University|
|Abstract:||The identification of fault-proneness at an early stage of software development is an imperative aspect to consider in order to reduce efforts in maintaining the quality of software products. newline newlineIn the development of a fault prediction model, combination of metrics results in better explanatory power of the model. The metrics used in combination are often interrelated, and do not have an additive effect, therefore the impact of a metric on another i.e. interaction need to be taken into account. newline newlineWe statistically establish the relevance of acknowledging the interaction between metrics. Such an interaction based modeling results into a significant increase in the explanatory power of the corresponding predictive model. newline newlineHowever, this further gives rise to the issue of handling an increased number of predictors and evolved nonlinearity due to complex interaction among metrics. newline newlineStatistical and rule based techniques have been explored to efficiently handle such situations, wherein the dependency between explanatory and response variables is more complicated than a simple linear additive relationship. newline newlineThis work contributes towards the development of an efficient predictive model involving interaction among predictive variables with an abbreviated set of in-fluential terms.|
|Appears in Departments:||University School of Information and Communication Technology|
Files in This Item:
|rinkaj goyal usict 2007.pdf||Attached File||1.97 MB||Adobe PDF||View/Open|
Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.