ABSTRACT
This chapter aims to propose a model for estimating the functional product size of a software development project based on project information known at the beginning of its life cycle. This model is based on selected data mining techniques, i.e., artificial neural networks and decision trees, as well as uses International Software Benchmarking Standards Group (ISBSG) benchmarking data. The estimated functional size of the software is the necessary basis for proper planning of software project costs. This limits the usually huge losses of funds invested in unsuccessful and often very expensive software projects. The SEMMA methodology was used to build the model. The accuracy of the proposed model is approximately 85%. It proves that the model can support the estimation of the software functional size and costs in development projects. However, the sensitivity index of the presented model is an area that needs further work.
