Machine Learning Based Method for Mining Anomaly Features of Network Multi Source Data

Aiming at the problem that the current network multi-source data anomaly diagnosis is not effective, this paper proposes a method of network multi-source data anomaly feature mining based on machine learning. First of all, a multi-source data feature recognition model is built based on the multi-level structure of machine learning. Then, the network multi-source data feature classification algorithm is designed and optimized to identify and locate the abnormal data features based on the classification results. Finally, the network multi-source data abnormal data screening model is constructed to mine the abnormal characteristics. The experimental results show that this method has high practicability and accuracy, and fully meets the research requirements.