基于变异算子改进蚁群算法学习的模糊认知图

传统的模糊认知图学习算法需要依靠专家知识来建立模糊认知图模型。在专家知识受到限制的一些领域,完全通过数据来学习模糊认知图的算法是有必要的。提出的基于蚁群优化的学习算法,可以有效做到通过历史数据学习模糊认知图;为避免蚁群算法陷入局部最优的问题;以及其学习所得模型不够精确的问题,引入变异算子以改进蚁群算法,来提高算法性能,避免局部最优问题,得到更准确的模型。最后,算法性能在人工数据上进行了测试。实验表明,算法性能良好,可以得到准确的模糊认知图模型。

Learning fuzzy cognitive maps by ant conoly algorithm improved by mutation operator

Traditionally FCMs are developed by experts.We need to learn FCMs directly from data when expert knowledge is limited in some areas.An algorithm based on ant conoly optimization is proposed to learn FCMs from data effectively in this paper.To avoid the problem of local optimization,mutation operator is lead up to improve the performance of ant conoly algorithm and obtain more accurate FCM model. Finally,the performance of the algorithm is tested on artificial data.The experiment shows that the algorithm has good performance and accurate FCM model can be obtained.