Rank thresholds in classifier ensembles in medical diagnosis

• Autorzy: Antczak K.

Warianty tytułu

PL

Progowanie rang dla zespołów klasyfikatorów w diagnostyce medycznej

• Języki publikacji: EN

Abstrakty

EN

Classification methods have multiple applications, with medical diagnosis being one of the most common. A powerful way to improve classification quality is to combine single classifiers into an ensemble. One of the approaches for creating such ensembles is to combine class rankings from base classifiers. In this paper, two rank-based ensemble methods are studied: Highest Rank and Borda Count. Furthermore, the effect of applying class rank threshold to these methods is analyzed. We performed tests using real-life medical data. It turns out that specificity of data domain can affect classification quality depending on classifier type.

Metody klasyfikacji mają wiele zastosowań, z których jednym z częściej spotykanych jest diagnostyka medyczna. Jakość klasyfikacji można w znaczący sposób podnieść, tworząc zespoły klasyfikatorów. Jedną z metod tworzenia takich zespołów jest łączenie rankingów generowanych przez klasyfikatory bazowe. W niniejszej pracy przeanalizowano dwie metody łączenia klasyfikatorów bazujące na rankingach: Najwyższej Rangi oraz Głosowanie Bordy. Dodatkowo zbadano wpływ progowania rankingu na jakość klasyfikacji. Testy przeprowadzono z użyciem rzeczywistych danych medycznych. Wykazano przy tym, że specyfika danych medycznych może wpłynąć na jakość klasyfikacji w zależności od typu klasyfikatora.

Słowa kluczowe

EN

medical diagnostics; classifier ensemble; rank threshold

PL

diagnostyka medyczna; łączenie klasyfikatorów; progowanie rankingu

• Wydawca: Institute of Computer and Information Systems, Faculty of Cybernetics, Military University of Technology
• Czasopismo: Computer Science and Mathematical Modelling
• Rocznik: 2016
• Tom: No. 3
• Strony: 5--11
• Opis fizyczny: Bibliogr. 25 poz., tab., wykr.

Twórcy

autor

Antczak K.

• Military University of Technology, Faculty of Cybernetics Institute of Computer and Information Systems Kaliskiego Str. 2, 00-908 Warsaw, Poland

Bibliografia

• [1] Tan P-N., Steinbach M., Kumar V., “Chapter 4. Classification: Basic Concepts, Decision Trees and Model Evaluation”, in: Introduction to Data Mining, 145–205, Pearson Education, 2006.
• [2] Wyatt J., Spiegelhalter D., “Field trials of medical decision-aids: potential problems and solutions”, Proceedings Annual Symposium Computer Applications in Medical Care, 3–7, AMIA, 1991.
• [3] Berger S.A., “GIDEON: A Computer Program for Diagnosis, Simulation, and Informatics in the Fields of Geographic Medicine and Emerging Diseases”, Emerging Infectious Diseases, Vol. 7, No. 7 (2001).
• [4] Oniśko A., Druzdzel M.J., Wasyluk H,. “Extension of the Hepar II Model to Multiple-Disorder Diagnosis”, in: Intelligent Information Systems, 303–313, Physica-Verlag HD, 2000.
• [5] Ramnarayan P., Tomlinson A., Kulkarni G., Rao A., Britto J., “A novel diagnostic aid (ISABEL): development and preliminary evaluation of clinical performance”, in: MEDINFO 2004, Marius Fieschi, Enrico Coiera, Yu-Chan Jack Li (Eds.), of the series: Studies in Health Technology and Informatics, Vol. 107, 1091–1095, IOS Press, 2004.
• [6] Walczak A., Gaj A., Jahnz-Różyk K., Wybrane zagadnienia informatycznego wspomagania decyzji medycznych, Andrzej Walczak (Ed.), Warszawa, Wojskowa Akademia Techniczna, 2013.
• [7] Rosenblatt F., The Perceptron: A Perceiving and Recognizing Automaton (Project PARA), New York, 1957.
• [8] Rish I., “An empirical study of the naive Bayes classifier”, Proceedings of IJCAI 2001 workshop on Empirical Methods in Artificial Intelligence, 3: 41–46, IBM New York, 2001.
• [9] Cortes C., Corinna C., Vladimir V., “Support-Vector Networks”, Machine Learning, 20, 273–297 (1995).
• [10] Yuan G-X., Guo-Xun Y., Chia-Hua H., Chih-Jen L., “Recent Advances of Large-Scale Linear Classification”, Proceeding of the IEEE, Vol. 100, 2584–2603, 2012.
• [11] Bland R., Learning XOR: Exploring the Space of a Classic Problem, University of Stirling, Department of Computing Sciennce and Mathematics, 1998.
• [12] Rosenblatt F., Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms, Spartan Books, Washington, 1962.
• [13] Neapolitan R.E., “Probabilistic Reasoning in Expert Systems: Theory and Algorithms”, Technometrics, Vol. 34, No. 1, 99–100 (1992).
• [14] Subasi M., Subasi E., Anthony M., Hammer P.L., “Using a similarity measure for credible classification”, Discrete Applied Mathematics, 157(5), 1104–1112. (2009).
• [15] Kuncheva L.I., “Switching between selection and fusion in combining classifiers: An experiment”, IEEE Transactions on Systems, Man, and Cybernetics – Part B: Cybernetics, Vol. 32, No. 2, 146–156 (2002).
• [16] Dietterich T.G., “Ensemble Methods in Machine Learning”, in: Multiple Classifier Systems 2000, J. Kittler and F. Roli (Eds.), of series Lecture Notes in Computer Science, Vol. 1857, 1–15, Springer-Verlag, Berlin, 2000.
• [17] Bradley A.P., “The use of the area under the ROC curve in the evaluation of machine learning algorithms”, Pattern Recognition, Vol. 30, No. 7, 1145–1159 (1997).
• [18] Xu L., Krzyzak A., Suen C.Y., “Methods of combining multiple classifiers and their applications to handwriting recognition”, IEEE Transactions on Systems, Man, and Cybernetics, Vol. 22, No. 3, 418–435 (1992).
• [19] Chen J.J., Tsai C-A., Moon H., Ahn H., Young J.J., Chen C-H., “Decision threshold adjustment in class prediction”, SAR QSAR Environmental Research, Vol. 17, No. 3, 337–352 (2006).
• [20] Mohammadi L., van de Geer S., “On threshold-based classification rules”, in: Mathematical Statistics and Applications: Festschrift for Constance van Eeden, Marc Moore, Sorana Froda and Christian Léger (Eds.), IMS Lecture Notes – Monograph Series, Vol. 42, 261–280, IMS, USA, 2003.
• [21] Koford J., Groner G., “The use of an adaptive threshold element to design a linear optimal pattern classifier”, IEEE Transaction on Information Theory, Vol. 12, 42–50 (1966).
• [22] Levitin G., “Threshold optimization for weighted voting classifiers”, Naval Research Logistics, Vol. 50(4), 322–344 (2003).
• [23] Tin Kam Ho., Ho T.K., Hull J.J., Srihari S.N., “Decision combination in multiple classifier systems”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 16, 66–75 (1994).
• [24] McLean I.S., McMillan A., Monroe B.L. (Eds.), “Partial Justification of the Borda Count”, in: The Theory of Committees and Elections by Duncan Black and Committee Decisions with Complementary Valuation by Duncan Black and RA Newing, 369–385, Springer Netherlands, 1998.
• [25] Walczak A., Paczkowski M., “Medical data preprocessing for increased selectivity of diagnosis”, in: Bio-Algorithms and Med- -Systems, Vol. 12(1), 39–43 (2016).

Uwagi

PL

Opracowanie ze środków MNiSW w ramach umowy 812/P-DUN/2016 na działalność upowszechniającą naukę.