Server risk checking method, computer device and computer program product

Abstract

The embodiment of the invention discloses a server risk checking method, a computer device and a computer program product. The method comprises the following steps: collecting historical log data of a plurality of server operation environments in multiple dimensions and respectively summarizing the historical log data into data subsets; preprocessing each data subset to generate a feature vector data set; dividing the plurality of server operation environment areas into a source domain and a target domain based on the data mark number of each feature vector data set; performing dynamic sampling on the feature vector data set according to a feature distribution difference value between the source domain and the target domain to generate a task sampling set; obtaining a detection model based on the task sampling set and the feature vector data set; and performing risk detection based on the to-be-detected log data of the target server and the detection model, and executing a response measure when determining that the target server has a risk. According to the method, the generalization of the model can be improved by combining multi-dimensional data acquisition and domain sampling with meta-learning, accurate cross-domain risk detection is realized, and the operation safety of the server is guaranteed.

Images