In this paper, we are interested in symbiotic radio networks (SRNs) and focus on the user association problem in SRNs. Specifically, in an SRN, the base station serves multiple cellular users using time division multiple access (TDMA) and each IoT device is associated with one cellular user for information transmission. The objective of user association is to link each IoT device to an appropriate cellular user by maximizing the sum rate of all IoT devices. However, the difficulty in obtaining the full real-time channel information makes it difficult to design an optimal policy for this problem.To overcome this issue, we propose a deep reinforcement learning (DRL) algorithm, which uses historical knowledge to infer the current information in order to make appropriate decisions for this user association problem. Finally, simulation results show that the proposed DRL algorithm achieves performance comparable to the optimal user association policy which requires perfect real-time information.