TY -的A2 -孟,鲎盟——刘Chunbo AU -潘,兰兰AU - Gu,昭君AU -王,古时的非盟,任通盟——王,智PY - 2020 DA - 2020/11/16 TI -有效概率异常检测模型系统日志SP - 8827185六世- 2020 AB -系统日志可以记录系统状态和系统操作中重要事件的细节。检测系统中异常日志是一种常见的现代大型分布式系统的方法。然而基于阈值的分类模型用于异常检测输出只有两个值:正常或异常,缺少的概率估计预测结果是否正确。本文采用统计学习算法Venn-Abers预测评估领域的预测结果的信心系统日志异常检测。它能够计算标签为一组样本的概率分布,并提供一个质量评估在一定程度上预测的标签。两个Venn-Abers预测LR-VA和SVM-VA基于逻辑回归和支持向量机实现,分别。然后,不同算法之间的差异被认为是,建立一个由叠加multimodel融合算法。然后Venn-Abers预测基于叠加算法叫做Stacking-VA实施。四种算法的性能(unimodel Venn-Abers预测基于unimodel multimodel,和Venn-Abers预测基于multimodel)比较的有效性和准确性。实验进行一个日志数据集的Hadoop分布式文件系统(HDFS)。 For the comparative experiments on unimodels, the results show that the validities of LR-VA and SVM-VA are better than those of the two corresponding underlying models. Compared with the underlying model, the accuracy of the SVM-VA predictor is better than that of LR-VA predictor, and more significantly, the recall rate increases from 81% to 94%. In the case of experiments on multiple models, the algorithm based on Stacking multimodel fusion is significantly superior to the underlying classifier. The average accuracy of Stacking-VA is larger than 0.95, which is more stable than the prediction results of LR-VA and SVM-VA. Experimental results show that the Venn-Abers predictor is a flexible tool that can make accurate and valid probability predictions in the field of system log anomaly detection. SN - 1530-8669 UR - https://doi.org/10.1155/2020/8827185 DO - 10.1155/2020/8827185 JF - Wireless Communications and Mobile Computing PB - Hindawi KW - ER -