KPI data dimension reduction method based on inter-metric and time dimension selection, electronic equipment and storage medium

The invention discloses a KPI data dimension reduction method based on inter-metric and time dimension selection, electronic equipment and a storage medium, and belongs to the technical field of anomaly detection processing. In order to improve precision and efficiency in multi-dimensional KPI anomaly detection, the method comprises the following steps: performing low-variance filtering processing on a multi-dimensional KPI data set to obtain a preprocessed multi-dimensional KPI data set; inter-metric dimension selection processing is carried out, a time sequence is used for coding and compressing a KPI sequence, then mean shift clustering is used for retaining correlation dimensions, and a multi-dimensional KPI data set of inter-metric dimension selection is obtained; performing time dimension selection processing on the preprocessed multi-dimensional KPI data set, including using 3-sigma to label an outlier of KPI data of each dimension, and then using sample entropy to select the dimension of the outlier c