Content analysis in information flows
The paper deals with architecture of content recognition system. To analyze the problem the stochastic model of content recognition in information flows was built. We proved that under certain conditions it is possible to solve correctly a part of the problem with probability 1, viewing a finite section of the information flow. That means that good architecture consists of two steps. The first step determines correctly certain subsets of contents, while the second step may demand much more time for true decision.
97 MATHEMATICS AND COMPUTING↗