User information collection method and device based on reinforcement learning model

The embodiment of the invention provides a user information collection method and device based on a reinforcement learning model, the reinforcement learning model comprises a strategy network, the method comprises the following steps: in a conversation process with a target user, a current environment state is acquired, the current environment state is determined at least based on corresponding previous N rounds of conversation content, and each round of conversation content comprises N rounds of conversation content; historical information collection questions and historical user feedback of the target user; the current environment state is input into a strategy network, Q values corresponding to all alternative information collection questions in an alternative question set in the current environment state are obtained, the alternative question set is determined based on historical user feedback and a preset knowledge base, and the preset knowledge base comprises the mapping relation between specified types