基于知识图谱的可视化推荐方法

In this paper, aiming at the limitations of existing visual recommendation methods in terms of personalization, accuracy and interpretability, we propose an innovative knowledge graph-based visualization recommendation method, which utilizes a structured knowledge graph to deeply understand the relationship between data features and chart types, and expects to improve the accuracy of visualization recommendation. Through in-depth analysis and feature extraction on a large-scale corpus, a knowledge graph covering data features, data columns and chart types is constructed and optimized using RotatE embedding learning method to ensure the accuracy and interpretability of recommendation results. Test results conducted on public datasets show that the method performs better than similar methods in the visualization recommendation task, significantly improves the accuracy and diversity of recommendations, enhances the interpretability of recommendation process.