Bayesian minimum mean-square error estimation for classification error - Part i: Definition and the bayesian mmse error estimator for discrete classification

IEEE Transactions on Signal Processing

Volumn 59, Issue 1, 2011, Pages 115-129

  Dalton, Lori A ^a   Dougherty, Edward R ^b,c  

^a Texas AandM University   (United States)

^b TRANSLATIONAL GENOMICS RESEARCH INSTITUTE   (United States)

^c UNIVERSITY OF TEXAS MD ANDERSON CANCER CENTER   (United States)

Author keywords

Bayesian estimation; classification; discrete histogram rule; error estimation; genomics; minimum mean square estimation; small samples

Indexed keywords

BAYESIAN ESTIMATIONS; DISCRETE HISTOGRAMS; GENOMICS; MINIMUM MEAN-SQUARE ESTIMATIONS; SMALL SAMPLES;

BAYESIAN NETWORKS; CLASSIFICATION (OF INFORMATION); COVARIANCE MATRIX; ERROR ANALYSIS; ERRORS; MEAN SQUARE ERROR; OPTIMIZATION; SIGNAL PROCESSING;

ESTIMATION;

EID: 79959465085     PISSN: 1053587X     EISSN: None     Source Type: Journal    
DOI: 10.1109/TSP.2010.2084572     Document Type: Article

References (48)

1. I. Shmulevich and E. R. Dougherty, Genomic Signal Processing. Princeton, NJ: Princeton Univ. Press, 2001.
2. E. R. Dougherty, "On the epistemological crisis in genomics," Current Genom., vol. 9, no. 2, pp. 69-79, 2008.
3. O. Persson, M. Krogh, L. H. Saal, E. Englund, J. Liu, R. Parsons, N. Mandahl, A. Borg, B. Widegren, and L. G. Salford, "Microarray anal- ysis of gliomas reveals chromosomal position-associated gene expres- sion patterns and identies potential immunotherapy targets," J. Neu- rooncol., vol. 85, pp. 11-24, 2007.
4. A. Guirguis, E. Elishaev, S. H. Oh, G. C. Tseng, K. Zorn, and J. A. DeLoia, "Use of gene expression proles to stage concurrent endometrioid tumors of the endometrium and ovary," Gynecol. Oncol., vol. 108, pp. 370-376, 2008.
5. T. J. Kim, J. J. Choi, W. Y. Kim, C. H. Choi, J. W. Lee, D. S. Bae, D. S. Son, J. Kim, B. K. Park, G. Ahn, E. Y. Cho, and B. G. Kim, "Gene expression proling for the prediction of lymph node metastasis in patients with cervical cancer," Cancer Sci., vol. 99, pp. 31-38, 2008.
6. B. S. Stolf, M. M. Santos, D. F. Simao, J. P. Diaz, E. B. Cristo, R. Hi- rata, M. P. Curado, E. J. Neves, L. P. Kowalski, and A. F. Carvalho, "Class distinction between follicular adenomas and follicular carci- nomas of the thyroid gland on the basis of their signature expression," Cancer, vol. 106, pp. 1891-1900, 2006.
7. A. Barrier, A. Lemoine, P. Y. Boelle, C. Tse, D. Brault, F. Chiappini, J. Breittschneider, F. Lacaine, S. Houry, M. Huguier, M. J. Van der Laan, T. Speed, B. Debuire, A. Flahault, and S. Dudoit, "Colon cancer prognosis prediction by gene expression proling," Oncogene, vol. 24, pp. 6155-6164, 2005.
8. F. De Smet, N. L. Pochet, K. Engelen, T. Van Gorp, P. Van Hummelen, K. Marchal, F. Amant, D. Timmerman, B. L. De Moor, and I. B. Ver- gote, "Predicting the clinical behavior of ovarian cancer from gene ex- pression proles," Int. J. Gynecol. Cancer, vol. 16, no. Suppl. 1, pp. 147-151, 2006.
9. A. Marchet, S. Mocellin, C. Belluco, A. Ambrosi, F. DeMarchi, E. Mammano, M. Digito, A. Leon, A. D'Arrigo, M. Lise, and D. Nitti, " Geneexpressionproleofprimarygastriccancer:Towardsthepredic- tion of lymph node status," Ann. Surg. Oncol., vol. 14, pp. 1058-1064, 2007.
10. E. Karlsson, U. Delle, A. Danielsson, B. Olsson, F. Abel, P. Karlsson, and K. Helou, "Gene expression variation to predict 10-year survival in lymph-node-negative breast cancer," BMC Cancer, vol. 8, p. 254, 2008.
11. M.Heuser,L.U.Wingen,D.Steinemann,G.Cario,N.vonNeuhoff,M. Tauscher,L.Bullinger,J.Krauter,G.Heil,H.Dhner,B.Schlegelberger, and A. Ganser, "Gene-expression proles and their association with drug resistance in adult acute myeloid leukemia," Haematologica, vol. 90, pp. 1484-1492, 2005.
12. M. La, E. H. Ahn, G. E. Mercado, S. Chuai, M. Edgar, B. R. Pawel, A. Olshen, F. G. Barr, and M. Ladanyi, "Global gene expression proling of PAX-FKHR fusion-positive alveolar and PAX-FKHR fusion-negative embryonal rhabdomyosarcomas," J. Pathol., vol. 212, pp. 143-151, 2007.
13. A. F. Ziober, K. R. Patel, F. Alawi, P. Gimotty, R. S. Weber, M. M. Feldman, A. A. Chalian, G. S. Weinstein, J. Hunt, and B. L. Ziober, "Identication of a gene signature for rapid screening of oral squamous cell carcinoma," Clin. Cancer Res., vol. 12, pp. 5960-5971, 2006.
14. G. N. Fuller, C. Mircean, I. Tabus, E. Taylor, R. Sawaya, J. M. Bruner, I. Shmulevich, and W. Zhang, "Molecular voting for glioma classica- tion reecting heterogeneity in the continuum of cancer progression," Oncol. Rep., vol. 14, pp. 651-656, 2005.
15. A.Barrier,F.Roser,P.Y.Bolle,B.Franc,C.Tse,D.Brault,F.Lacaine, S. Houry, P. Callard, C. Penna, B. Debuire, A. Flahault, S. Dudoit, and A. Lemoine, "Prognosis of stage ii colon cancer by non-neoplastic mucosa gene expression proling," Oncogene, vol. 26, pp. 2642-2648, 2007.
16. N. Glick, "Additive estimators for probabilites of correct classica- tion," Pattern Recognit., vol. 10, no. 3, pp. 211-222, Jan. 1978.
17. U. M. Braga-Neto and E. R. Dougherty, "Is cross-validation valid for small-sample microarray classication," Bioinformatics, vol. 20, no. 3, pp. 374-380, 2004.
18. B. Hanczar,J. Hua, and E. R. Dougherty, "Decorrelation of the true and estimated classier errors in high-dimensional settings," EURASIP J. Bioinform. Syst. Biol., vol. 2007, Jan. 2007, Article ID 38473, 12 pp.
19. U. Braga-Neto and E. R. Dougherty, "Bolstered error estimation," Pat- tern Recognit., vol. 37, no. 6, pp. 1267-1281, Jun. 2004.
20. C. Sima and E. R. Dougherty, "Optimal convex error estimators for classication," Pattern Recognit., vol. 39, no. 6, pp. 1763-1780, 2006.
21. A. Wald, "On a statistical problem arising in the classicaion of an individual into one of two groups," Ann. Math. Statist., vol. 15, no. 2, pp. 145-162, Jun. 1944.
22. R. Sitgreaves, "Some results on the distribution of the W-classica- tion statistic," in Studies in Item Analysis and Prediction, H. Solomon, Ed. Stanford, CA: Stanford Univ. Press, Jan. 1961, pp. 241-251.
23. U. Braga-Neto and E. R. Dougherty, "Exact performance of error es- timators for discrete classiers," Pattern Recognit., vol. 38, no. 11, pp. 1799-1814, Nov. 2005.
24. A. Zollanvari, U. M. Braga-Neto, and E. R. Dougherty, "On the sam- pling distribution of resubstitution and leave-one-out error estimators forlinearclassiers,"PatternRecognit.,vol.42,no.11,pp.2705-2723, Nov. 2009.
25. A. Zollanvari, U. M. Braga-Neto, and E. R. Dougherty, "On the joint sampling distribution between the actual classication error and the resubstitution and leave-one-out error estimators for linear classiers," IEEE Trans. Inf. Theory, vol. 56, no. 2, pp. 784-804, 2010.
26. M. J. Sorum, "Estimating the Probability of Misclassication," Ph.D. dissertation, Univ. of Minnesota, Minneapolis, Dec. 1968.
27. S. Geisser, "Estimation associated with linear discriminants," Ann. Math. Statist., vol. 38, no. 3, pp. 807-817, Jun. 1967.
28. E. T. Jaynes, "Prior probabilities," IEEE Trans. Syst. Sci. Cybern., vol. 4, no. 3, pp. 227-241, Sep. 1968.
29. J. O. Berger, Statistical Decision Theory and Bayesian Analysis, 2nd ed. New York: Springer-Verlag, 1985.
30. M. H. DeGroot, Optimal Statistical Decisions. New York: McGraw- Hill, 1970.
31. H. Akaike, "The interpretation of improper prior distributions as limits of data dependent properprior distributions," J.Roy. Statist.Soc. Series B, Methodol., vol. 42, no. 1, pp. 46-52, 1980.
32. J. L. Devore, Probability and Statistics for Engineering and the Sci- ences, 4th ed. Pacic Grove, CA: Brooks/Cole, 1995.
33. L. Devroye, L. Gyr, and G. Lugosi, A Probabilistic Theory of Pattern Recognition,ser.StochasticModellingandAppliedProbability. New York: Springer-Verlag, 1996.
34. N. Glick, "Sample-based multinomial classication," Biometrics, vol. 29, no. 2, pp. 241-256, 1973.
35. M. Goldstein and W. R.Dillon, DiscreteDiscriminant Analysis. New York: Wiley, 1978.
36. M. Hills, "Discrimination and allocation with discrete data," J. Roy. Statist. Soc. Series C, Appl. Statist., vol. 16, no. 3, pp. 237-250, 1967.
37. U. Braga-Neto and E. R. Dougherty, "Exact correlation between actual and estimated errors in discrete classication," Pattern Recognit. Lett., vol. 31, pp. 407-413, 2010.
38. U. Braga-Neto, "Classication and error estimation for discrete data," Current Genom., vol. 10, no. 7, pp. 446-462, Nov. 2009.
39. V. P. Kuznetsov, "Stable detection when the signal and spectrum of normal noise are inaccurately known," Telecommun. Radio Eng., vol. 30-31, pp. 58-64, 1976.
40. S. A. Kassam and T. I. Lim, "Robust wiener lters," J. Franklin Inst., vol. 304, no. 415, pp. 171-185, Oct./Nov. 1977.
41. A. M. Grigoryan and E. R. Dougherty, "Design and analysis of robust optimal binary lters in the context of a prior distribution for the states of nature," Math. Imag. Vis., vol. 11, no. 3, pp. 239-254, Dec. 1999.
42. A. M. Grigoryan and E. R. Dougherty, "Bayesian robust optimal linear lters," Signal Process., vol. 81, no. 12, pp. 2503-2521, Dec. 2001.
43. E. R. Dougherty, J. Hua, Z. Xiong, and Y. Chen, "Optimal robust clas- siers," Pattern Recognit., vol. 38, no. 10, pp. 1520-1532, 2005.
44. R. Pal, A. Datta, and E. R. Dougherty, "Bayesian robustness in the control of gene regulatory networks," IEEE Trans. Signal Process., vol. 57, no. 9, pp. 3667-3678, Sep. 2009.
45. A. Lunts and V. Brailovsky, "Evaluation of attributes obtained in sta- tistical decision rules," Eng. Cybern., vol. 3, pp. 98-109, 1967.
46. M. Stone, "Cross-validatory choice and assessment of statistical pre- dictions," J. Roy. Statist. Soc., vol. 36, pp. 111-147, 1974.
47. B. Efron, "Bootstrap methods: Another look at the jackknife," Ann. Statist., vol. 7, pp. 1-26, 1979.
48. B. Efron, "Estimating the error rate of a prediction rule: Improvement on cross validation," J. Amer. Statist. Assoc., vol. 78, pp. 316-331, 1983.