Network Anomaly Data Mining Method Based on Association Rule Improvement
Traditional methods for mining network anomaly data suffer from inaccuracies when calculating the distance between the network anomaly data and the associated centers, resulting in limited mining precision. So, this research proposes an improved network anomaly data mining method based on association rule algorithms. The method begins by initializing association centers for network anomaly data, followed by employing local search and optimization for network anomaly entities using the K-means algorithm; then, it accurately calculates the distance between network anomaly data and association centers using association rule algorithms;finally, it identifies the network anomalies that are furthest from the association centers to complete the mining process. Research results indicate that, when mining the same quantity of network anomaly data, this method significantly increases the proportion of accurately mined network anomalies compared to traditional methods, enhancing precise identification of network anomaly data, and having significant advantages.
association rule algorithmnetwork anomalydata mining analysis
万方数据
维普
