%0 Conference Paper %1 Kultur:2008:ESE:1453101.1453148 %A Kultur, Yigit %A Turhan, Burak %A Bener, Ayse Basar %B Proceedings of the 16th ACM SIGSOFT International Symposium on Foundations of Software Engineering %C New York, NY, USA %D 2008 %I ACM %K myown %P 330--338 %R 10.1145/1453101.1453148 %T ENNA: Software Effort Estimation Using Ensemble of Neural Networks with Associative Memory %U http://doi.acm.org/10.1145/1453101.1453148 %X Companies usually have limited amount of data for effort estimation. Machine learning methods have been preferred over parametric models due to their flexibility to calibrate the model for the available data. On the other hand, as machine learning methods become more complex they need more data to learn from. Therefore the challenge is to increase the performance of the algorithm when there is limited data. In this research we used a relatively complex machine learning algorithm, neural networks, and showed that stable and accurate estimations are achievable with an ensemble using associative memory. Our experimental results revealed that our proposed algorithm (ENNA) achieves on the average PRED(25) = 36.4 which is a significant increase compared to Neural Network (NN) PRED(25) = 8. %@ 978-1-59593-995-1

@inproceedings{Kultur:2008:ESE:1453101.1453148, abstract = {Companies usually have limited amount of data for effort estimation. Machine learning methods have been preferred over parametric models due to their flexibility to calibrate the model for the available data. On the other hand, as machine learning methods become more complex they need more data to learn from. Therefore the challenge is to increase the performance of the algorithm when there is limited data. In this research we used a relatively complex machine learning algorithm, neural networks, and showed that stable and accurate estimations are achievable with an ensemble using associative memory. Our experimental results revealed that our proposed algorithm (ENNA) achieves on the average PRED(25) = 36.4 which is a significant increase compared to Neural Network (NN) PRED(25) = 8.}, acmid = {1453148}, added-at = {2015-09-17T22:25:04.000+0200}, address = {New York, NY, USA}, author = {Kultur, Yigit and Turhan, Burak and Bener, Ayse Basar}, biburl = {https://www.bibsonomy.org/bibtex/2395b57656c36780fe65880bd13a6ed24/burak.turhan}, booktitle = {Proceedings of the 16th ACM SIGSOFT International Symposium on Foundations of Software Engineering}, description = {ENNA}, doi = {10.1145/1453101.1453148}, interhash = {ab24086e4081c1d8f77b9c9d1980ada0}, intrahash = {395b57656c36780fe65880bd13a6ed24}, isbn = {978-1-59593-995-1}, keywords = {myown}, location = {Atlanta, Georgia}, numpages = {9}, pages = {330--338}, publisher = {ACM}, series = {SIGSOFT '08/FSE-16}, timestamp = {2015-09-17T22:25:04.000+0200}, title = {ENNA: Software Effort Estimation Using Ensemble of Neural Networks with Associative Memory}, url = {http://doi.acm.org/10.1145/1453101.1453148}, year = 2008 }