中文摘要：

针对IaaS（Infrastructure as a Service）云平台中用户异常行为的检测问题,提出了一种基于用户行为模型和神经网络相结合的异常检测方法.该方法通过构造一种基于时间、地点和事件的用户行为模型,在此基础上建立用户的正常行为模式,并与神经网络算法相结合,将用户当前行为网络输出值与给定阈值进行比较,以此来判断用户的行为是否异常,从而实现用户行为的异常检测.实验结果表明,相比其它类似的用户行为检测方法,该方法能更有效发现用户的异常行为.

英文摘要：

To detect users＇ abnormal behavior in IaaS cloud platform, an anomaly detection method based on users＇ behavior and neural network was proposed. In order to correctly detect users＇ abnormal behaviour, a users＇ behavioral model was constructed based on time, place and event, and on this basis, users＇ normal behavior patterns were established. By combining the model and neural network, the method can detect and identify whether the behavior is normal or not by comparing the network output of the users＇ current behavior with a given threshold. The experimental resuhs show that the proposed method can detect users＇ abnormal behavior more effectively than other similar methods of users＇ behaviour detection.