Neural Network Models for Agile Software Effort Estimation based on Story Points

摘要: Agile software development is now accepted as a superior alternative to conventional methods of development, because its inherent benefits like iterative rapid delivery and reduced risk. Hence, the industry must be able efficiently estimate effort necessary develop projects using agile methodology. For this, different techniques expert opinion, analogy, disaggregation etc. are adopted by researchers practitioners. But no proper mathematical model exists for this. The existing ad-hoc thus prone incorrect. One popular approach calculating mathematically Story Point Approach (SPA). In this study, an has been made improve prediction accuracy estimation done SPA. doing types neural networks (General Regression Neural Network (GRNN), Group Method Data Handling (GMDH) Polynomial Cascade- Correlation Network) used. Finally, performance models generated these compared analyzed. story points. team defines relationship between point effort. Usually 1 equal ideal working day. Total no. points that can convey in sprint (an iteration development) called ``team velocity" or per sprint. SPA, aggregate number alongside velocity project considered determine Now obtaining better accuracy, three used study. results obtained applying their accessed