Enhancement of prediction accuracy in COCOMO model for software project using neural network

In this research, it is concerned with constructing software effort estimation model based on artificial neural networks. The model is designed accordingly to improve the performance of the network that suits to the COCOMO model. In this paper, it is proposed to use multi layer feed forward neural network to accommodate the model and its parameters to estimate software development effort. The network is trained with back propagation learning algorithm by iteratively processing a set of training samples and comparing the network´s prediction with the actual effort. COCOMO dataset is used to train and to test the network and it was observed that proposed neural network model improves the estimation accuracy of the model. The test results from the trained neural network are compared with that of the COCOMO model. The preliminary results obtained suggest that the proposed architecture can be replicated for accurately forecasting the software development effort. The aim of this study is to enhance the estimation accuracy of COCOMO model, so that the estimated effort is more close to the actual effort.