From Function Points to COSMIC - A Transfer Learning Approach for Effort Estimation

Abstract

Software companies exploit data about completed projects to estimate the development effort required for new projects. Software size is one of the most important information used to this end. However, different methods for sizing software exist and companies may require to migrate to a new method at a certain point. In this case, in order to exploit historical data they need to resize the past projects with the new method. Besides to be expensive, resizing is also often not possible due to the lack of adequate documentation. To support size measurement migration, we propose a transfer learning approach that allows to avoid resizing and is able to estimate the effort of new projects based on the combined use of data about past projects measured with the previous measurement method and projects measured with the new one. To assess our proposal, an empirical analysis is carried out using an industrial dataset of 25 projects. Function Point Analysis and COSMIC are the measurement methods taken into account in the study

Proceedings of the 17th International Conference on Predictive Models and Data Analytics in Software Engineering

“…This is probably linked to the potential increase in the level of heterogeneity of datasets when using CC training projects [32]. Approaches that attempt to handle heterogeneity have shown more successful results in improving predictive performance over WC approaches [11,23,48,49]. However, these approaches assume that heterogeneity only affects input features, lacking mechanisms to cope with concept drift.…”

Definition and evaluation of a COSMIC measurement procedure for sizing Web applications in a model-driven development environment

“…The effort estimation models built for this particular company may also be useful for other similar companies who do not have historical data on effort estimation. Nevertheless, the results obtained should be further validated using larger datasets including data from other software companies [88,89,90,91,92,93], and in the future we therefore plan to apply OO-HCFP in other contexts and also consider larger Web project datasets.…”