User Behavior Prediction and Interface Personalization Design Combined with Deep Q-Network

User behavior prediction and interface personalized design of computer software are important research directions to improve children's learning effect and user experience. The traditional children's art education software often adopts the unified interface design and teaching methods, which cannot meet the personalized needs of different users. Based on the deep Q network (Deep Q-Network, DQN) algorithm, the software predicts the user behavior, and designs the personalized interface for it. By collecting and analyzing the user behavior data of children's art education software, the deep Q network model is constructed. Then, the model is used to predict future user behavior, thus providing personalized learning recommendations and interface design for the software. Finally, the validity and feasibility of this method are verified by experiments. The experimental results show that the prediction accuracy of this method can reach 94.7%, thus effectively helping parents to make the correct choice and use. This paper provides an effective computer method for user behavior prediction and interface personalized design of children's art education software, which is of great significance to improving the quality of children's art education.