An Unsupervised Model on Outsourcing Software Projects Risks Based on T-S Fuzzy Neural Network

DOI

10.2991/eame-15.2015.114How to use a DOI?

Keywords

outsourcing software projects risks; unsupervised model; semi-supervised model; T-S fuzzy neural network; standard score

Abstract

Supervised learning algorithm is widely used in current research of software project risks, while it cannot evaluate the risk without decision class. Considering this, we present an unsupervised model, which can solve the forecasting problem of outsourcing software projects risks in unsupervised situation, based on T-S fuzzy neural network (T-S FNN) and the concept of standard score in risk evaluation of outsourcing software project. In simulation, we use 200 samples to train and test the network repeatedly, helping us to get the accuracy rate of 76.2%. After that, the standard score of 0.558 is applied to the forecasting of 60 testing samples, and the accuracy of them is 75%, which shows the stability of the model. Moreover, the comparison with some supervised neural networks shows that the forecasting accuracy of T-S FNN is close to that of the supervised models above. Meanwhile, T-S FNN shows a more balanceable forecasting performance in both successful project and failure project.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Z.Z. Ouyang
AU  - J.Z. Chen
AU  - Z.H. Zhang
AU  - Y. Hu
PY  - 2015/07
DA  - 2015/07
TI  - An Unsupervised Model on Outsourcing Software Projects Risks Based on T-S Fuzzy Neural Network
BT  - Proceedings of the 2015 International Conference on Electrical, Automation and Mechanical Engineering
PB  - Atlantis Press
SP  - 405
EP  - 408
SN  - 2352-5401
UR  - https://doi.org/10.2991/eame-15.2015.114
DO  - 10.2991/eame-15.2015.114
ID  - Ouyang2015/07
ER  -