Model of the situations recognition in conditions dissimilar and incomplete data

Abstract

There have been proposed model of situations recognition in determined alphabet, based on combination of quantitative and qualitative characteristics, considered the data dissimilar, coming from information sources. In the model improved the quality of the formalized description of quantitative characteristics using histograms instead of fuzzy L-R intervals, which gives the possibility of organizing a self-learning recognition system situations based on the processing statistics of recognition. Recommendations for formation of characteristics sets to overcome the data incompleteness have been present. Proposed model may be use for formalization knowledge about situation recognition process. Subject areas for implementation taken results are diagnostic in medicine and energetic, situation assessment at the military control points, making prognosis in economic.