规划识别是一种根据观察数据识别和推断被观察对象目的或意图的预测理论 .在计算机系统入侵检测研究中 ,为了提前预测出异常事件的发生 ,提出了一种基于规划识别理论的入侵企图预测方法 .通过对主机上的系统调用序列为观察对象建立预测模型 ,提出了一种带参数补偿的贝叶斯网络动态更新算法 ,对观察对象的目的进行预测 .实验结果表明动态贝叶斯网络对预测系统调用序列中的异常入侵企图有较高的精度 . Planning recognition is a kind of prediction theory which identifies and infers the purpose or intention of the observed object based on the observed data.In the study of computer system intrusion detection, in order to predict the occurrence of abnormal events in advance, an intrusion attempt prediction based on the planning recognition theory A dynamic update algorithm of Bayesian network with parameter compensation is proposed to predict the purpose of the observed object.Experimental results show that dynamic Bayesian network pair Predicting anomalous intrusion in a system call sequence attempts to have higher precision.