Security policy selection method and device based on LSTM terminal adaptive optimization, and medium

The invention discloses a security policy selection method and device based on LSTM (Long Short Term Memory) terminal adaptive optimization, and a medium, and the method comprises the steps: combining a threat type, an occurrence probability, a system vulnerability level, a system security level and time through LSTM, forming the recognition and analysis of an attack intention, and finally generating a maximum possible attack path. And a security policy model is selected based on the attack loss, corresponding cost and security level requirements of the nodes related to the attack path. Therefore, according to the embodiment of the invention, on the basis of pre-judging the attack intention, the terminal access security risk is measured by adopting the maximum possibility attack path, the dynamic change of the terminal risk and the system security requirement can be quickly adapted, and in addition, the resource dynamics is also considered, so that the security risk of the terminal can be effectively reduced.