To solve the problems of the poor convergence speed and being easy to fall into the local optimum in the particle swarm algorithm, an attribute reduction algorithm based on genetic quantum particle swarm (GQPSO) is presented. GQPSO takes advantage of the wide search range of quantum system and utilizes the selection and variation of the genetic algorithm to avoid algorithm premature convergence local optimum and get considerable convergence speed. The experiment shows that GQPSO has a faster convergence rate and global search capabilities, which improves the efficiency of the attribute reduction.