Electrical and Computer Engineering Publications

Document Type


Publication Date


URL with Digital Object Identifier



In this paper, we propose a novel Artificial Neural Network (ANN) to predict software effort from use case diagrams based on the Use Case Point (UCP) model. The inputs of this model are software size, productivity and complexity, while the output is the predicted software effort. A multiple linear regression model with three independent variables (same inputs of the ANN) and one dependent variable (effort) is also introduced. Our data repository contains 240 data points in which, 214 are industrial and 26 are educational projects. Both the regression and ANN models were trained using 168 data points and tested using 72 data points. The ANN model was evaluated using the MMER and PRED criteria against the regression model, as well as the UCP model that estimates effort from use cases. Results show that the ANN model is a competitive model with respect to other regression models and can be used as an alternative to predict software effort based on the UCP method.

Citation of this paper:

@inproceedings{DBLP:conf/icmla/NassifCH12, author = {Ali Bou Nassif and Luiz Fernando Capretz and Danny Ho}, title = {Estimating Software Effort Using an ANN Model Based on Use Case Points}, booktitle = {ICMLA (2)}, year = {2012}, pages = {42-47}, ee = {http://dx.doi.org/10.1109/ICMLA.2012.138}, crossref = {DBLP:conf/icmla/2012-2}, bibsource = {DBLP, http://dblp.uni-trier.de} } @proceedings{DBLP:conf/icmla/2012-2, title = {11th International Conference on Machine Learning and Applications, ICMLA, Boca Raton, FL, USA, December 12-15, 2012. Volume 2}, booktitle = {ICMLA (2)}, publisher = {IEEE}, year = {2012}, isbn = {978-1-4673-4651-1}, ee = {http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6403616}, bibsource = {DBLP, http://dblp.uni-trier.de} }

Find in your library