UML Based Object Oriented Software Development Effort Estimation Using ANN

Abstract

Today the software industry may be the fastest growing industries in the world. As the industry grows the size and cost of the software are also increasing. Hence there is need of effective techniques for cost estimation in order to control the costs and make the software more competitive. The software development techniques are also changing, and now- a- days most of the companies using object – oriented techniques for software development. UML diagrams are used for analysis and design of a software and different UML constructs can be used at different stages of software development for estimation of resources like efforts & cost etc. Use case point and class point analysis were accepted for the estimation of object-oriented software. In class point analysis some unknown constants are multiplied with known values of product attributes, like no. of methods, classes etc. These add impurities to the estimation model. So in this paper we have applied Artificial Neural model for estimation directly from known attributes and we find that it gives good results than class point analysis.