Large-scale high-dimensional high-speed streaming data online anomaly detection method and system

The invention belongs to the technical field of streaming data mining, and particularly relates to a large-scale high-dimensional high-speed streaming data online anomaly detection method and system, and the method comprises the steps: employing a matrix sketch model to carry out the processing of a data block transmitted at a high speed, and obtaining a sketch matrix; inputting the sketch matrix into a hash learning model to obtain an optimal model parameter and a feature hash table at the current moment; constructing an abnormal score calculation model according to the optimal model parameters and the feature hash table, inputting to-be-detected sample data into the abnormal score calculation model for detection, and judging whether the sample data is abnormal or not. By adopting the matrix sketch and Hash learning technology, the data scale and the feature dimension can be reduced, the detection speed and the storage efficiency can be improved, and the dynamic change of stream data distribution can be self-adapted by updating the detection model on line; the problem that anomaly detection cannot be efficiently carried out on the streaming data in real time in the current large-scale high-dimensional and high-speed environment is effectively solved.