Using Decision-Making Block of Computer-Based Intelligent Biomedical Avatar for Applied Research in Bioinformatics

Abstract

A biomedical task in which the definitions and properties of applied research indicators under study in bioinformatics is formalized. A wide range of traditional approaches used for predicting medical time series were reviewed. Advanced algorithms for predicting moments of reversals of biomedical trends based on machine learning tools were investigated as well. The effectiveness of different kinds of approaches was discussed, and related examples are given. An original securities price dynamics trend classification algorithm, based on the use of the sliding window methodology and biomedical avatar, is described. A general scheme of the classification algorithm to identify biomedical market phases is analyzed and results of computer modelling are presented. Selection of initial and resulting metrics is grounded.