Towards an early software estimation using log-linear regression and a multilayer perceptron model
Journal of Systems and Software
Hi-index | 0.00 |
Software effort estimation is one of the most important tasks in software engineering. Software developers conduct software estimation in the early stages of the software life cycle to derive the required cost and schedule for a project. In the requirements stage, where most software estimation is conducted, the available information is usually imprecise or incomplete. In this paper, a new regression model is created for software effort estimation based on use case point model. Furthermore, a Sugeno Fuzzy Inference System (FIS) approach is applied on this model to improve the estimation. Results show that an improvement of 11 % can be achieved in MMRE after applying the Sugeno fuzzy logic approach.