Aiming at the high-dimensional data generated by intrusion detection system and Extreme Learning Machine (ELM) parameter optimization problems, a model based on Improved Artificial Bee Colony (IABC) combined with ELM is proposed. At first the ABC algorithm is applied to select input layer weight and hidden layer bias of the ELM, and the root mean square error of each individual in the initial population is taken as fitness degree value. In order to further improve calculation accuracy of the algorithm, the algorithm adopts the strategy of self-adaptive adjustment neighbor search step length. A...
