T. V. Zakharova, S. E. Kinzhitaeva, “Method of weighted discriminant systems for the classification of objects with missing data”, Sistemy i Sredstva Inform., 2016, Volume 26, Issue 4,Pages <nobr>89

Method of weighted discriminant systems for the classification of objects with missing data

T. V. Zakharova^ab, S. E. Kinzhitaeva^c

^a Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
^b Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
^c Moscow Institute of Physics and Technology (State University), 9 Institutskiy Per., Dolgoprudny, Moscow Region 141700, Russian Federation

Abstract: The authors describe a method of classification of objects with missing or partially known data. The tasks related to the processing of incomplete data are common in medicine. Patient data can contain gaps or missing. Information classification of patients with varying degrees of schizophrenia was carried out using the new method. Schizophrenia is a genetic disease; so, important is the task of studying the genetic predisposition of a person to the disease. Analysis of associations between polymorphisms of genes was performed. A distinctive feature of the provision of medical data is its emptiness by more than 70%. The dimension of discriminant signs was significantly reduced and high reliability of forecasting has been received.

Keywords: discriminant analysis; censored data; classification functions; genetic diseases.

Received: 15.09.2016

DOI: 10.14357/08696527160408