Software effort estimation through clustering techniques of RBFN network

Now a day’s software cost/effort estimation is a very complex job to do. Several estimation techniques have been developed in this regard. This assessment of parameters like, time, cost, and number of staff required sequentially which in turn is to be done at an early stage. Constructive Cost model which is also known as COCOMO model was one of the best model to estimate the cost and time in person month of a software project. The estimation of cost and time supports the project planning and tracking, as well as also controls the expenses of software development process. The accurate effort estimation will lead to improve the project success rate. In this paper, the main focus is on finding the accuracy of estimation of effort/cost of software using radial basis function neural network (RBFN) incorporating ANN-COCOMO II which can be used for functional approximation. This model estimates the total effort of software development according to the characteristics of COCOMO-II along with radial basis clustering techniques. The RBFN network is much faster than other network because the learning process in this network has two stages and both stages can be made efficient by appropriate learning algorithms. The RBFN network uses COCOMO-II dataset for training.