Optimization of project effort estimate using neural network

Software project effort estimate is an important means to ensure software development. Software estimation is an integral, costly, time consuming activity in the software development life cycle. It involves running the system being developed under a variety of circumstances and configurations. As the volume and complexity of software applications are continuously increasing, software estimate becomes very effort intensive task and is prone to errors. Hence, accurate estimation is not only save time and resources when applications are updated or developed but also speed up the process of update or development. Estimation is more than a set of estimate runs to generate apparent results. It includes accurate cost, effort, time, benefit monitoring, interpreting a broad range of result etc. Considering all these aspects it has been the intention to develop a generalized user friendly estimation environment for software development, where any software can be estimated with minimal manual interaction. This paper presents a performance analysis of artificial neural network. Furthermore it is the intention to minimize the magnitude of relative errors between actual effort and calculated effort.