The embodiment of the invention discloses a user identity recognition method and device, and relates to the technical field of data processing, and the method comprises the steps: collecting historical behavior data of a user, the historical behavior data at least comprising social mode behavior data, dialogue mode data and consumption mode data; Generating an identity label of a user according tothe historical behavior data; And identifying whether the user identity is legal according to the identity. The identity identification of the user is generated according to the non-biological characteristics of the user, the legality of the user identity is identified by using the identity identification, and the performance of user identity identification is improved.

A user identity identification method and device

The invention provides a mobile social network user identity falseness identification method based on behavioral modeling to judge user legitimacy by judging consistence of user behaviors. The mobilesocial network user identity falseness identification method based on behavioral modeling benefits from rich information data provided by the current mobile social network, and mass user social data including position, moving mode, social relationship, user-generated content, shopping records and the like is acquired. Multivariate data is taken as a study object, and features of the information are extracted. The user behaviors are modeled in the three following dimensions with mixed kernel density estimation, LDA text modeling and other methods: (1) geographical location, (2) user-generated text and (3) social relationship, blocking rate, interrupt rate and precision ratio of detection are calculated by simulating a user identity theft test, system performance is evaluated on the basis, problems of the traditional identification system are solved, and a new idea and a new analytical method are provided for solving security problems in the information age.

Yang Bo

Papers

Mobile social network user identity falseness identification method based on behavioral modeling