Abstract
Artificial Neural Networks provide models for a large class of natural and artificial phenomena that are difficult to handle using classical parametric techniques. They offer a potential solution to fit all the data, including any outliers, instead of removing them. This paper compares the predictive performance of linear and nonlinear models in outlier detection. The best-subsets regression algorithm for the selection of minimum variables in a linear regression model is used by removing predictors that are irrelevant to the task to be learned. Then, the ANN is trained by the Multi-Layer Perceptron to improve the classification and prediction of the linear model based on standard nonlinear functions which are inherent in ANNs. Comparison of linear and nonlinear models was carried out by analyzing the Receiver Operating Characteristic curves in terms of accuracy and misclassification rates for linear and nonlinear models. The results for linear and nonlinear models achieved 68% and 93%, respectively, with better fit for the nonlinear model.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Landi, A., Piaggi, P., Laurino, M., Menicucci, D.: Artificial neural networks for nonlinear regression and classification. In: 10th International Conference on Intelligent Systems Design and Application (ISDA), pp. 115–120 (2010)
Rusiecki, A.: Robust LTS backpropagation learning algorithm. In: Sandoval, F., Prieto, A., Cabestany, J., Graña, M. (eds.) IWANN 2007. LNCS, vol. 4507, pp. 102–109. Springer, Heidelberg (2007). doi:10.1007/978-3-540-73007-1_13
Lima, A.R., Cannon, A.J., Hsieh, W.W.: Nonlinear regression in environmental sciences by support vector machines combined with evolutionary strategy. Comput. Geosci. 50, 136–144 (2012)
Gujarati, D.N., Porter, C.: Basic Econometrics. McGraw Hill Education, Singapore (2009)
Williams, G., Baxter, R., He, H., Hawkins, S., Gu, L.: A comparative study of RNN for outlier detection in data mining. CSIRO Math. Inf. Sci. 1, 709–712 (2002). ISSN: 0–7695-1754-4
Motulsky, H.J., Brown, R.E.: Detecting outliers when fitting data with nonlinear regression - a new method based on robust nonlinear regression and the false discovery rate. BMC Bioinform. 7(123), 20 (2006)
Han, H.-G., Wang, L.-D., Qiao, J.-F.: Efficient self-organizing multilayer neural network for nonlinear system modeling. Neural Netw. 43, 22–32 (2013)
Muzhou, H., Lee, M.H.: A new constructive method to optimize neural network architecture and generalization, 1–8 (2013). CoRR abs/1302.0324
Garces, H., Sbarbaro, D.: Outliers detection in industrial databases: an example sulphur recovery process. In: World Congress, vol. 18(1) (2011)
Singh, K., Upadhyaya, S.: Outlier detection: applications and techniques. Int. J. Comput. Sci. Issues (IJCSI) 9(3), 307–323 (2012)
Lichman, M.: UCI Machine Learning Repository. University of California, Irvine (2013)
Khashei, M., Hamadani, A.Z., Bijari, M.: A novel hybrid classification model of ANN and Multiple linear regression models. Expert Syst. Appl. 39(3), 2696–2720 (2012)
Kutner, M.H., Nachtsheim, C.J., Neter, J.: Applied Linear Regression Models. McGraw Hill, New York (2008)
Fallah, N., Gu, H., Mohammed, K., Seyyedsalehi, S.A., Nourijelyani, K., Eshraghian, M.R.: Nonlinear poisson regression using neural networks: simulation study. Neural Comput. Appl. 18(8), 939–943 (2009)
Husin, N.A., Salim, N.: A comparative study for backpropagation neural network and nonlinear regression models for predicting dengue outbreak. Junal Teknologi Maklumat Bil 20(4), 97–112 (2008)
Maliki, O.S., Agbo, A.O., Maliki, A.O., Ibeh, L.M., Agwu, C.O.: Comparison of regression model and artificial neural network model for the prediction of electrical power generated in Nigeria. Adv. Appl. Sci. Res. 2(5), 329–339 (2011). ISSN: 0976–8610
Yang, P., Zhu, Q., Zhong, X.: Subtractive clustering based RBF neural network model for outlier detection. J. Comput. 4(8), 755–761 (2009)
Liu. Q., Lu, J., Chen, S., and Zhao, K.: Multiple naïve bayes classifiers ensemble for traffic incident detection. Math. Probl. Eng. 2014 (2014)
Tiryaki, S., Aydin, A.: An ANN for predicting compression strength of heat treated woods and comparison with a multiple linear regression model. Constr. Build. Mater. 62, 102–108 (2014)
Cateni, S., Colla, V., Vannucci, M.: Outlier detection methods for industrial applications. INTECH Open Access Publishers (2008)
Koncsos, T.: The application of neural networks for solving complex optimization problems in modeling. In: Conference of Junior Researchers in Civil Engineering (2012)
Cherkassky, V., Mulier, F.M.: Learning from Data: Concepts, Theory and Methods, 2nd edn., pp. 1538–1550 (2007). ISSN-13: 978–0471681823
Bo, Z.: A prediction model based on linear regression and artificial neural network analysis of the hairiness of polyester cotton winding yarn. Adv. Multimedia Softw. Eng. Comput. 128, 97–103 (2012)
Acknowledgments
This project is sponsored by Universiti Tun Hussein Onn Malaysia under the Short Term Grant (STG) Scheme Vot U129.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Dalatu, P.I., Fitrianto, A., Mustapha, A. (2017). A Comparative Study of Linear and Nonlinear Regression Models for Outlier Detection. In: Herawan, T., Ghazali, R., Nawi, N.M., Deris, M.M. (eds) Recent Advances on Soft Computing and Data Mining. SCDM 2016. Advances in Intelligent Systems and Computing, vol 549. Springer, Cham. https://doi.org/10.1007/978-3-319-51281-5_32
Download citation
DOI: https://doi.org/10.1007/978-3-319-51281-5_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-51279-2
Online ISBN: 978-3-319-51281-5
eBook Packages: EngineeringEngineering (R0)