Receiver Operating Characteristic for Diagnosis of Wine Quality by Bayesian Network Classifiers

Abstract:

This paper is dedicated to demonstrate the use of the receiver operating characteristic (ROC) and the area under the ROC curve (AUC) for diagnosing forecast skill. Several local search heuristic algorithms to discover which one performs better for learning a certain Bayesian networks (BN). Five heuristic search algorithms, including K2, Hill Climbing, Repeated Hill Climber, LAGD Hill Climbing, and TAN, were empirically evaluated and compared. This study tests BN models in a real-world case, the Vinho Verde wine taste preferences. An average AUC of 0.746 and 0.727 respectively in red wine and white wine were obtained by TAN algorithm. The results show that the use of TAN can effectively improve the AUC measures for predicting quality grade.

• Periodical: Advanced Materials Research (Volumes 591-593)
• Pages: 1168-1173
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.591-593.1168
• Online since: November 2012
• Authors: Chih Chiang Wei
• Keywords: Bayesian Network (BN), Classification, Receiver Operating Characteristic

References

[1] M.H. Zweig, G.N. Petrovich and Z.M Prijovich: Clinical Chemistry Vol. 39 (1993), p.1345

[2] M. Greiner, D. Pfeiffer, R.D. Smith: Preventive Veterinary Medicine Vol. 45 (2000), p.23

[3] Wikipedia: http://en.m.wikipedia.org/wiki/Receiver_operating_characteristic/ (2012)

[4] G. Vanagas: Interactive CardioVasc Thoracic Surgery Vol. 3 (2004), p.319

[5] C.E. Metz: Journal of the American College of Radiology Vol. 3 (2006), p.413

[6] I. Kurt, M. Ture and A.T. Kurum: Expert Systems with Applications Vol. 34 (2008), p.366

[7] A. Frank and A., Asuncion: UCI Machine Learning Repository, http://archive.ics.uci.edu/ml. School of Information and Computer Science, University of California, Irvine (2010)

[8] U.B. Kjaerulff and A.L. Madsen: Bayesian networks and influence diagrams. A guide to construction and analysis. Springer, New York (2008)

[9] F. Rijmen: International Journal of Approximate Reasoning Vol. 48 (2008) p.659

[10] J. Sun: Neural Networks Vol. 23 (2010), p.334

[11] M.A. Gomez-Villegas, P. Main, H. Navarro and R. Susi: Expert Systems with Applications Vol. 38 (2011), p.12409

[12] T.A. Lasko, J.G. Bhagwat, K.H. Zou and L. Ohno-Machado: Journal of Biomedical Informatics Vol. 38 (2005), p.404

[13] J.A. Hanley and B.J. McNeil: Radiology Vol. 143 (1982), p.29

[14] P. Cortez, A. Cerdeira, F. Almeida, T. Matos and J. Reis: Decision Support Systems Vol. 47 (2009), p.547

DOI: 10.1016/j.dss.2009.05.016

[15] CVRVV: Portuguese Wine: Vinho Verde, Comissao de Viticultura da Regiao dos Vinhos Verdes (CVRVV), http://www.vinhoverde.pt (2012)

DOI: 10.18803/capsi.v17.334-344

[16] A. Gopnik and J.B. Tenenbaum: Developmental Science Vol. 10 (2007), p.281

[17] F.V. Jensen: An introduction to Bayesian networks. UCL Press Limited, London (1996)

[18] F. Pernkopf: Pattern Recognition Vol. 38 (2005), p.1

[19] E. Saheli and R. Gras: Learning and Intelligent Optimization. IEEE conference (2009)

[20] G. Cooper and E. Herskovits: Machine Learning Vol. 9 (1992), p.309

[21] I.H. Witten and E. Frank: Practical machine learning tools and techniques with Java implementations. Morgan Kaufmann, San Francisco (2000)

[22] W.L. Buntine: IEEE Transactions on Knowledge and Data Engineering Vol. 8 (1996), p.195

[23] N. Friedman, D. Geiger and M. Goldszmidt: Machine Learning Vol. 29 (1997), p.131

[24] R.R. Bouckaert, E. Frank, M. Hall, R. Kirkby, P. Reutemann, A. Seewald and D. Scuse: WEKA Manual for Version 3-7-3. University of Waikato, New Zealand (2010)

[25] J.A. Swets: Science Vol. 240 (1988), p.1285