Multi-Layered Algorithm for Chinese Cultural Dissemination Power Based on Feature Selection and Weight Learning

This paper presents an innovative approach to analyzing and enhancing the effectiveness of cultural dissemination using a multi-layered algorithm framework based on feature selection and weight learning. We first employ the Least Absolute Shrinkage and Selection Operator (Lasso) regularization technique for feature selection, identifying the most informative features crucial for predicting the power of cultural transmission. Following this, a reinforcement learning framework based on Deep Q-Networks (DQN) is established, incorporating a reward mechanism that favors feature combinations promoting cultural dissemination. Through interaction with the environment, the model learns the weights of these features, reflecting their contribution to successful cultural transmission. The identified features and learned weights are then integrated into a multi-layered algorithmic framework. Each layer of this framework represents a different aspect of cultural transmission, such as content creation, dissemination channels, and audience feedback, ensuring effective interaction between layers. Finally, the model is applied to real-world cultural dissemination cases, like popular music, movies, or literary works, to validate its effectiveness. The results demonstrate the potential of this approach in providing insightful strategies for optimizing cultural dissemination.