Neural network based static prediction of software reliability in integrated navigation system
Jia Zhang · Ha'erbin gongye daxue xuebao · 2002
Accurate prediction of the number of faults in program modules is a major problem for quality control of a large scale integrated navigation software system. According to software test record and software complexity, a neural network was used as a tool for prediction of the number of faults in a program on a trial basis. It is indicated that the neural network did produce the models with better quality of fit and predictive quality than traditional linearity regression method.