Improved Bayesian Network and Its Application in Autonomous Capability Evaluation

The Bayesian network (BN) is a new intelligent evaluation technology, which has become one of the effective methods to solve uncertainty problems and deal with asymmetric information. However, its application limited in case complex indicator systems interrelated variables. To these problems, an improved BN model based on game theory proposed. First, IHORAFA attribute reduction algorithm used optimize index system. Then, weighted proposed for problem correlation among indicators, uses combination weighting method determine optimal weight improve accuracy calculation. Finally, applied autonomous capability ground attack UAVs. simulation results show that can be assessment reasoning under uncertain conditions variable correlation.