Intelligent level grade evaluation method and system based on restricted Boltzmann machine

The invention discloses an intelligent level grade evaluation method and system based on a restricted Boltzmann machine, and relates to the technical field of artificial intelligence evaluation, and the method comprises the steps: firstly obtaining redisk data obtained by a plurality of tested agents in pairwise fight in a test task, and then for each tested agent, carrying out the redisk data; the method comprises the following steps of: obtaining a plurality of evaluation indexes through calculation according to battle data of a tested agent in each piece of redisk data corresponding to the tested agent, taking the plurality of evaluation indexes as input, utilizing a trained restricted Boltzmann machine network to predict to obtain ELO integral of the tested agent, and obtaining a final ELO integral according to an average value of the ELO integral of multi-disk data. And finally, for each tested agent, the intelligent level grade of the tested agent is determined according to the final ELO integral of the