Scheme Selection of Software Quality Based on Support Vector Machine

Abstract

Software Quality scheme selection is an important pre-work of software project, For software quality evaluation index are difficult to quantify and has the characteristics of uncertainty, this paper uses a nonlinear SVM classification algorithm research. Firstly, to build a software quality scheme evaluation index system composed of 12 indicators, and through hierarchical structure analysis model representation; Then, research on support vector machine, using graphic to show that the basic idea, through mathematical model to describe the algorithm, constructed Sigmoid kernel function; Finally, to perform case analysis, through 8 have been evaluated schemes to train, drawn 4 unselected scheme evaluation results. This article will apply SVM to software quality scheme selection, it is SVM a new exploration in the field of software engineering, and to improve software quality has an important significance.