Prediction of software maintainability using fuzzy logic

Abstract

The relationship between object oriented metrics and software maintainability is complex and non-linear. Therefore, there is considerable research interest in development and application of sophisticated techniques which can be used to build models for predicting software maintainability. However, when predicting maintainability not only product quality measurements are surrounded with imprecision and uncertainty, but also the relationships between the external and internal quality attributes suffer from imprecision and uncertainty. The reason behind that, there are at least two important sources of information for building the prediction model: historical data and human experts. Therefore, in this paper an attempt has been made to utilize the capability of fuzzy logic in handling imprecision and uncertainty to come up with an efficient maintainability prediction model. The proposed model is constructed using object-oriented metrics data in Li and Henry's datasets collected from two different object-oriented systems.