Simulation on potential risk mining model of underlying network dormant data
Jin Wang, WanRu Chen · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2015
This paper focuses on the potential risk mining method of underlying network dormant data, a potential risk mining method based on bat algorithm optimization BP neural network is put forward.The BP neural network parameter is coded as the individual of bat, and the accuracy of the potential risk mining is viewed as the individual fitness function, then through the simulation of bat flight process to find the optimal parameters of BP neural network, finally according to the optimal parameters to establish the potential risk mining model of underlying network dormant data.The experimental results show that, this algorithm can effectively improve the mining efficiency, and further reduce the error.