Classification of Dangerous Situations for Small Sample Size Problem in Maintenance Decision Support Systems

P. 338–345.

Milov V., Savchenko A.

In this paper we examine the maintenance decision support in classification of the dangerous situations discovered by the monitoring system. This task is reduced to the contextual multi-armed bandit problem. We highlight small sample size problem appeared in this task due to the rather rare failures. The novel algorithm based on the nearest neighbor search is proposed. An experimental study is provided for several synthetic datasets with the situations described by either simple features or grayscale images. It is shown, that our algorithm outperforms the well-known contextual multi-armed methods with the Upper Confidence Bound and softmax random search strategies. 

Keywords: распознавание образов, методы поддержки принятия решения, pattern recognition, малые выборки, small sample size problems, contextual multi-armed bandit, контекстный многорукий бандит, decision support system

In book

5th Conference on Analysis of Images, Social Networks, and Text (AIST 2016)

Springer, 2017.