Alternative Methods to Rank the Impact of Object Oriented Metrics in Fault Prediction Modeling using Neural Networks

The aim of this paper is to rank the impact of Object Oriented(OO) metrics in fault prediction modeling using Artificial Neural Networks(ANNs). Past studies on empirical validation of object oriented metrics as fault predictors using ANNs have focused on the predictive quality of neural networks versus standard statistical techniques. In this empirical study we turn our attention to the capability of ANNs in ranking the impact of these explanatory metrics on fault proneness. In ANNs data analysis approach, there is no clear method of ranking the impact of individual metrics. Five ANN based techniques are studied which rank object oriented metrics in predicting fault proneness of classes. These techniques are i) overall connection weights method ii) Garson-s method iii) The partial derivatives methods iv) The Input Perturb method v) the classical stepwise methods. We develop and evaluate different prediction models based on the ranking of the metrics by the individual techniques. The models based on overall connection weights and partial derivatives methods have been found to be most accurate.

Authors:

• Kamaldeep Kaur
• Arvinder Kaur
• Ruchika Malhotra

Keywords:

• Artificial Neural Networks (ANNS)
• Backpropagation
• Fault Prediction Modeling.

References:

[1] S. R. Chidamber and C. F. Kemerer, "A Metrics Suite for Object
Oriented Design," IEEE Transactions on Software Engineering, vol. 20,
pp 476-493,1994
[2] V.R. Basili, et al,"A Validation of Object-Oriented Design Metrics as
Quality Indicators," IEEE Transactions on Software Engineering, vol.
22, pp 751-761,1996
[3] L.C. Briand , Jurgen Wust, "Modeling Development Effort in Object
Oriented Systems using Design Properties," IEEE Transactions on
Software Engineering, vol. 27, no. 11, 2001
[4] T.M. Khoshgoftaar, E.B.Allen, J.P. Hudephol and S.J. Aud," E.B.Allen,
J.P. Hudephol and S.J. Aud," ," Application of neural networks to
quality modeling of a very large telecommunication system", IEEE
Transactions on Neural Networks, vol.8,pp. 902-909,1997
[5] M. M. T. Thwin and T.-S. Quah,, "Application of Neural Networks for
predicting Software Development faults using Object Oriented Design
Metrics", Proceedings of the 9th International Conference on Neural
Information Processing, November 2002, pp. 2312 - 2316.
[6] M. M. T. Thwin and T.-S. Quah,,, "Prediction of Software Readiness
using Neural Networks," ICITA2002, ISBN:1-86467-114-9
[7] Tibor Gyimothy, Rudolf Fernec, and Istvan Siket , " Empirical Validaion
of Object -Oriented Metrics on Open Source Software for Fault
Prediction" , IEEE Transactions on Software Engineering, vol. 31, no.
10, October 2005
[8] K.K. Aggarwal, Y. Singh. P. Chandra,M.Puri, "Evaluation of various
training Algorithms for Sofware Engineering Applications," ACM
SIGSOFT Software Engineering Notes ,vol 30, no.4 July 2005.
[9] Dietrich Schusche, Chun Nan Hsu, Hung-Ju Huang , "A weight
analysis based wrapper approach to Neural Nets feature subset
selection," IEEE Transactions on Systems, Man and Cybernetics,
vol.32,pp. 207-21,2002
[10] G.D. Garson, "Interpreting Neural Network Connection Weights," AI
Expert 6, pp. 47-51, 1991.
[11] D.W. Ruck, S.K Rogers, M. Kabrisky, "Feature Selection using Multi
Layer Perceptrons," in Journal of Neural Network Computing, vol 2, no
. 2 , pp. 40-48 ,1990.
[12] S. Haykins, "A Comprehensive Foundation on Neural Networks ,"
Prentice Hall, 1999
[13] Khaled El Emam, "A Methodology for Validating Software Product
Metrics," National Research Council Canada, Institute for Information
Technology, ERB-1076.