面向对象软件的数据异常监测方法研究

Research on data exception monitoring method of object-oriented software

为了提高面向对象跨项目软件的可靠性,需要进行面向对象软件的数据异常监测,提出基于行为形态特征提取的面向对象软件数据异常监测方法。采用大数据度量方法进行面向对象跨项目软件的特征检测,结合融合特征分类方法进行面向对象跨项目软件的自适应形态学特征检测和差异性融合处理,提取面向对象跨项目软件的异常特征量,通过交叉编译映射的方法进行面向对象跨项目软件的异常数据监测和程序加载,构建面向对象跨项目软件的特征选择算法,通过优先级排序的方法进行面向对象跨项目软件数据的异常等级排序,结合多策略的特征筛选方法进行面向对象跨项目软件的行为形态特征提取,实现面向对象跨项目软件的数据异常监测优化。仿真结果表明,采用该方法进行面向对象跨项目软件异常监测的准确性较高,可以提升软件的可靠性和缺陷故障诊断能力。

In order to improve the reliability of object-oriented cross-project software,data anomaly monitoring of object-oriented software is needed,and a data anomaly monitoring method of object-oriented software based on behavioral morphological feature extraction is proposed.The feature detection of object-oriented cross-project software is carried out by using big data measurement method,the adaptive morphological feature detection and difference fusion processing of object-oriented cross-project software are carried out by combining fusion feature classification method,the abnormal feature quantity of object-oriented cross-project software is extracted,the abnormal data monitoring and program loading of object-oriented cross-project software are carried out by cross-compiling mapping method,and the feature selection algorithm of object-oriented cross-project software is constructed.The data of object-oriented cross-project software is ranked by the method of prioritization,and the behavioral morphological features of object-oriented cross-project software are extracted by the method of multi-strategy feature screening,so as to realize the data anomaly monitoring optimization of object-oriented cross-project software.The simulation results show that the accuracy of object-oriented cross project software anomaly monitoring is high,and the software reliability and fault diagnosis ability can be improved.