Procedures for improvement of quality of diagnostic advisor built on binary classifiers

Authors

  • Григорій Василійович Кіт Open International University of Human Development "Ukraine" Str. Khoreva 1, Kyiv, Ukraine, 04071, Ukraine
  • Володимир Анатолійович Павлов Open International University of Human Development "Ukraine" Str. Khoreva 1, Kyiv, Ukraine, 04071, Ukraine
  • Олександр Володимирович Павлов National Technical University of Ukraine "Kyiv Polytechnic Institute" Peremohy Ave, 37, Kyiv, Ukraine, 03056, Ukraine

DOI:

https://doi.org/10.15587/1729-4061.2013.9129

Keywords:

Binary classifiers, hemostasis, differential diagnosis, GMDH

Abstract

The work aims to improve the quality of classification of diagnostic advisor on the example of the problem of differential diagnosis of mild forms of hemostasis, such as von Willebrand's disease, coagulopathy, deaggregating thrombocytopathy, combined pathology of hemostasis. To solve the problems we have compared the efficiency of diagnostic advisor, developed by the relaxation iterative algorithm GMDH according to two approaches. In the first variant, the system was developed on binary classifiers, built on the principle of "one against all"; in the second the diagnostic problem was solved by a single classifier for four classes. For this task the advantages of the first variant were shown. The further step was the improvement of the system by introduction of procedures for solving classifiers conflicts. With this purpose we proposed to design additional functions of classification on data sets with different combinations of samples of classes that must be differentiated: classification of each pair of diagnoses, classification of diagnosis from a couple of other diagnoses, classification of diagnoses in pairs. Thus, a multi-classification system is formed, where at each further level the conflicts left unsolved at lower levels, are solved. This approach should be applied while formulating a criterion of diagnostic system as "the maximum number of correct diagnosis

Author Biographies

Григорій Василійович Кіт, Open International University of Human Development "Ukraine" Str. Khoreva 1, Kyiv, Ukraine, 04071

Кандидат технічних наук

Кафедра комп'ютерного еколого-економічного моніторингу

Володимир Анатолійович Павлов, Open International University of Human Development "Ukraine" Str. Khoreva 1, Kyiv, Ukraine, 04071

Кандидат технічних наук

Кафедра комп'ютерного еколого-економічного моніторингу

Олександр Володимирович Павлов, National Technical University of Ukraine "Kyiv Polytechnic Institute" Peremohy Ave, 37, Kyiv, Ukraine, 03056

Ph.D.

Departments of descriptive geometry, engineering and computer graphics

References

  1. Эдвард, А. Патрик. Основы теории распознавания образов [Текст]: пер. с англ. - М.:«Советское радио»,1980. - 407с.
  2. Павлов, А. В. Синтез классификаторов дифференциальной диагностики заболеваний легких форм гемостазиопатий методом группового учета аргументов [Текст] / Ан. В.Павлов, В. А. Павлов, В. В. Томилин // Восточно-Европейський журнал передовых технологий.–Харьков.- 2011. - № 2.- С. 54-65.
  3. Ивахненко, А.Г. Помехоустойчивость моделирования. [Текст] / А.Г.Ивахненко, В.С. Степашко. — К.: «Наук.думка»,1985. - 216 с.
  4. Павлов, А.В. Методика экспериментальных исследований сходимости итерационных алгоритмов метода группового учёта аргументов [Текст] / А. В. Павлов, В. А. Павлов // Вісник НТУУ „КПІ”. Інформатика, управління та обчислювальна техніка: Зб. наук. пр. – К., «Век+», – 2011. – №54, - С. 36-40.
  5. Павлов, А.В. О сходимости обобщенного релаксационного итерационного алгоритма метода группового учета аргументов [Текст] / А. В.Павлов, Н. В. Кондрашова // УСиМ. 2012. - №3. - С.24-38.

Published

2013-02-20

How to Cite

Кіт, Г. В., Павлов, В. А., & Павлов, О. В. (2013). Procedures for improvement of quality of diagnostic advisor built on binary classifiers. Eastern-European Journal of Enterprise Technologies, 1(3(61), 11–15. https://doi.org/10.15587/1729-4061.2013.9129

Issue

Section

Control systems