Neural Network model for software size estimation using Use Case Point approach

Abstract

A realistic software size estimation is required in various subject areas of Software Engineering. Especially to predict performance of software system using Software Performance Engineering approach, software size is an important input parameter. Even though several estimation procedures are available the Neural Network model presents advantages over normal estimation procedure. In this paper we develop a Neural Network model to estimate the size of software using Use Case Point approach. The results are validated and a case study of Multi-Agent System is presented.