Learning and adaptation in real-time decision support systems of a semiotic type

This paper describes the learning and adaptation methods for the real-time decision support systems (RTDSSs) of a semiotic type intended for operative-dispatching management of a complex object or a process. It is taken into consideration that RTDSSs are mostly oriented towards open and dynamic problem domains, where incompleteness and uncertainty of input information are present. This work was supported by the Russian Fund of Basic Research (project no. 02-07-90042).