An adaptive learning approach to software cost estimation

Software cost estimation predicts the amount of effort and development time required to build a software system. It is one of the most critical tasks and it helps the software industries to effectively manage their software development process. There are a number of cost estimation models. The most widely used model is Constructive Cost Model (COCOMO). In this paper, the use of back propagation neural networks for software cost estimation is proposed. The model is designed in such a manner that accommodates the COCOMO model and improves its performance. It also enhances the predictability of the software cost estimates. The model is tested using two datasets COCOMO dataset and COCOMO NASA 2 dataset. The test results from the trained neural network are compared with that of the COCOMO model. From the experimental results, it was concluded that the integration of the conventional COCOMO model and the neural network approach improves the cost estimation accuracy and the estimated cost can be very close to the actual cost.