With the rapid development of the smart cultural tourism industry, how to rationally allocate digital resources to improve the overall operational efficiency has become an urgent problem to be solved. In this paper, a deep reinforcement learning (DQN)-based digital resource allocation optimization model for cultural and tourism industry is proposed. The model estimates the Q-value function by deep neural network, which solves the resource allocation problem of cultural and tourism industry in a complex cloud computing environment. The experiments use the Google Trace dataset to simulate different sizes of cloud environments for task scheduling. The experimental results show that the proposed model significantly outperforms traditional algorithms in terms of task execution success rate and resource utilization. For example, in a cluster of 75 servers, the task execution success rate reaches 0.742, which is higher than DRL (0.593) and AIRL (0.646). In addition, the model exhibits a higher success rate when dealing with low latency tolerant tasks, proving its advantage in dealing with urgent task scheduling. The study shows that the application of the DQN-based resource allocation model in the cultural tourism industry effectively improves the resource utilization efficiency and system throughput capacity.
