Software Quality Prediction using Fuzzy Rule based System

Development of software quality, development effort etc is a common application of software metrics. A new tool as Fuzzy Logic offers a good technique for building models for software quality prediction. This paper illustrates the practice of estimation at a personal level using projects and presents the results obtained with a fuzzy rule based system and an ordinary multiple regression. Result shows that the value of MMRE applying fuzzy logic is smaller than MMRE applying multiple regression; while the value of Pred (0.25) and Pred (0.05) applying fuzzy logic is higher than Pred (0.25) and Pred (0.05) applying multiple regression. Thus Results demonstrate that fuzzy logic can be used as alternative for predicting the software quality.