基于大数据技术的计算机网络异常数据检测方法

Detection Method of Computer Network Abnormal Data Based on Big Data Technology

在传统的网络异常数据检测方法中,直接利用原始数据进行分析,导致检测结果的准确性相对较低,为此提出基于大数据技术的计算机网络异常数据检测方法。首先利用原始网络数据的均值和平均绝对偏差参数对数据信息进行标准化处理,在利用大数据技术提取了以误差模糊程度为指标的数据学习表征后,通过调整分类系数的取值结果,使得对应的检测精度满足不同环境的检测需求。当标准化网络数据的波动小于误差模糊系数时,利用此时的分类系数对待检测数据进行分类,并根据分类数据波动程度与误差模糊系数之间的关系对数据的状态作出判断。测试结果中,设计检测方法对不同类型异常数据的有效检出率达到了90.0%以上,明显优于对比方法。

In the traditional network abnormal data detection methods,the original data is directly used for analysis,resulting in relatively low accuracy of detection results.Therefore,a computer network abnormal data detection method based on big data technology is proposed.First,the mean value and average absolute deviation parameters of the original network data are used to standardize the data information.After using big data technology to extract the data learning representation with the error ambiguity degree as the indicator,the corresponding detection accuracy can meet the detection requirements of different environments by adjusting the value results of the classification coefficient.When the fluctuation of the standardized network data is less than the error ambiguity coefficient,The classification coefficient is used to classify the detected data,and the state of the data is judged according to the relationship between the fluctuation degree of the classified data and the error ambiguity coefficient.In the test results,the effective detection rate of the designed detection method for different types of abnormal data has reached more than 90.0%,which is obviously superior to the comparison method.