International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE)
Accurately estimating software effort is probably the biggest challenge facing software developers. Estimates done at the proposal stage has high degree of inaccuracy, where requirements for the scope are not defined to the lowest details, but as the project progresses and requirements are elaborated, accuracy and confidence on estimate increases. It is important to choose the right software effort estimation techniques for the prediction of software effort. Artificial Neural Network (ANN) and Support Vector Machine (SVM) have been used for prediction of software effort in this work. The performance indices Sum-Square-Error (SSE), Mean-Square-Error (MSE), Root-Mean-Square-Error (RMSE), Mean-Magnitude-Relative-Error (MMRE), Relative-Absolute-Error (RAE), Relative-Root-Square-Error (RRSE), Mean-Absolute-Error (MAE), Correlation Coefficient (CC) and PRED (25) have been used to compare the results obtained from these two methods.