MEASURING SOFTWARE COMPLEXITY USING NEURAL NETWORKS

Measuring the software complexity is an important task in the management of software projects. In the recent years, many researchers have paid much attention to this challenging task due to the commercial importance of software projects. In the literature, there are some software metrics and estimation models to measure the complexity of software. However, we still need to introduce novel models of software metrics to obtain more accurate results regarding software complexity. In this paper, we will show that neural networks can be used as an alternative method for estimation of software complexity metrics. We use a neural network of three layers with a single hidden layer and train this network by using distinct training algorithms to determine the accuracy of software complexity. We compare our results of software complexity obtained by using neural networks with those calculated by Halstead model. This comparison shows that the difference between our estimated results obtained by Bayesian Regularization Algorithm with 10 hidden neurons and Halstead calculated results of software complexity is less than 2%, implying the effectiveness of our proposed method of neural networks in estimating software complexity