A touch-screen user key-press behavior pattern construction and analysis system and an identity recognition method, wherein the method comprises: performing data analysis by using the history key-press information of password input via the soft keyboard on a touch screen of a mobile phone and establishing a corresponding neutral network model, and carrying out a model calculation for new data to be measured to identify the user's identity. The system consists of a user data collection module, a data preprocessing module, a model training module and a user identity authentication module, wherein, the user data collection module is used for collecting the time sequence information, pressure and the contact area information of the user's clicks on the soft keyboard for password input, the data preprocessing module is used for preprocessing the collected data, removing dirty data and normalizing the data, the model training module is used for analyzing and modeling of all users' key-press patterns, and the user identity authentication module is used for performing a model calculation on the new data to be measured so as to identify the user's identity and improve the security of user accounts and passwords.
展开▼