Client information query abnormal behavior detection method and device

The invention relates to a client information query abnormal behavior detection method and device. The client information query abnormal behavior detection method comprises the following steps: obtaining query data of client information queried by a user group; according to each query behavior characteristic in the query data, counting at least two groups of query behaviors of a single user in theuser group and corresponding frequencies; calculating a PCA score of data queried by a single user in each time aggregation unit to form a first data set; calculating the DTW distance between every two users according to the first data set to form a second data set; and dividing the user groups into a plurality of groups by adopting a clustering analysis method based on the second data group, andoutputting user information of which the number of people in the groups is less than a preset threshold. Abnormal people are identified from a group, the detection range is reduced, and the detectionaccuracy and the detection