Research on Power Communication Abnormal Data Extraction Based on the Improved Hadoop Mining Framework
Abnormal data of power communication system are often hidden in massive data,which leads to low coverage of Hadoop mining framework in abnormal data extraction.Therefore,the research on abnormal data extraction of power communication based on improved Hadoop mining framework is proposed.The data quality is improved by preprocessing strategies such as standardization,filtering and normalization of complex signals.Local data aggregation optimization components are introduced to optimize data transmission,multi-NameNode Hadoop architecture is adopted to solve the bottleneck problem of single node,and K-Means clustering algorithm is combined for data mining.Through feature evaluation and screening and parallel clustering analysis,the key abnormal data features are effectively identified.The experimental results show that this method can significantly improve the extraction coverage of abnormal data.
improving Hadoop mining frameworkpower communication systemanomalous datafeature extractioncluster analysis
万方数据
维普
