Xia, W., Capretz, L.F., Ho, D. Neuro-Fuzzy Approach to Calibrate Function Points, Fuzzy, Vancouver, June 2007.
[Abstract] Function Points is an important and well-accepted software size metric. However, it is absolutely essential to accurately calibrate Function Point (FP), whose aims are to fit specific software application, to reflect software industry trend, and to improve cost estimation. Neuro-Fuzzy is a technique that incorporates the learning ability from neural network and the ability to capture human knowledge from fuzzy logic. We developed a Neuro-Fuzzy model to calibrate Function Points. The empirical validation using ISBSG data repository Release 8 shows a 22% improvement in software effort estimation after calibration using Neuro-Fuzzy technique.
Please send request to info@nfa-estimation.com for the full paper.