Quantum Chicken Swarm Optimization with Levy Flight and Its Application in Parameter optimization of Random Forest

In order to solve the problem of unbalanced global and local search ability and optimization accuracy in chicken swarm optimization, we propose a quantum chicken swarm optimization with Levy flight (L-QCSO). First, the quantum system is used to increase the search space of the roosters. Second, the Levy flights strategy is used to improve the hens behavior. Finally, the improved chicken swarm optimization is used to optimize the number of tree and the number of split variables pre-selected by the tree node on the random forest. In the experiment, results show that the optimized chicken swarm optimization is effective and better than the chicken swarm optimization on the benchmark functions. In the end, the optimized chicken swarm optimization is used to optimize RF, which is applied into oil layer recognition. Results show that this optimization model is superior to the RF in terms of recognition rate and time.