大型云计算信息系统的异常数据检测模型仿真  

Simulation of Abnormal Data Detection Model for Large Scale Cloud Computing Information System

作　　者：杨静[1] 

机构地区：[1]内蒙古财经大学计算机系,内蒙古呼和浩特市010070

出　　处：《计算机仿真》2015年第11期378-381,共4页Computer Simulation

摘　　要：研究大型云计算信息系统的异常数据检测方法。采用传统方法进行大型云计算信息系统的异常数据检测,由于系统数据量较大且呈非线性分布,在进行异常数据检测时,容易产生收敛速度慢,检测准确率低的不足。为此,提出基于模糊隐马尔科夫模型的大型云计算信息系统的异常数据检测方法。依据相关理论对检测模型的模型系数、方差和转移矩阵等参数进行初始化,将隐马尔科夫检测方法与模糊理论进行结合,分别从三个角度进行数据状态判别,并计算检测结果正确的数据集的隶属度函数,最后进行隶属度合并,并据此对当前数据进行判断,从而实现对大型云计算信息系统的异常数据的有效检测。实验结果表明,采用改进算法进行大型云计算信息系统的异常数据检测,能够有效提高检测效率与有效性,能够满足实际检测需求。The detection method of abnormal data in large-scale cloud computing information system was researched. A detection method of abnormal data in large-scale cloud computing information system based on hidden markov model was proposed. According to the related theory, the model coefficient, variance, transfer matrix and other parameters of the detected model were initialized. The hidden Markov detection method was combined with fuzzy theory, the data state was judged from three angles respectively, and the membership function of the data set with cor- rected testing result was calculated. Finally, the membership was merged, and on the basis of current data, the judgment was made, so as to realize the effective detection of abnormal data in large-scale cloud computing information system. The experimental results show that the improved algorithm can effectively improve the detection efficiency and effectiveness and satisfy the requirements of actual detection.

关 键 词：大型云计算信息系统 异常数据检测 隐马尔科夫模型 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]