A Measurement Method of Software Flexibility Based on BP Network

An approach for software flexibility measurement and measurement indexes based on BP neural network were proposed. By using the self-organizing, self-learning and self- adapting properties of BP neural networks, the rule of software flexibility measurement indexes weight concealed in the training data could be learned by means of BP neural networks automatically adjust indexes weight of measurement. Therefore the influence of random factors could be overcome. In order to validate the proposed method, the software flexibility measurement model based on BP neural network was established. 10 samples were trained and 2 samples were tested. Analysis results show that we could get satisfying precision of the software flexibility measurement and the method is feasible in the software flexibility measurement.