Analysis of the Deepening Path of Cognitive Logic in Chinese Philosophy Based on Knowledge Mapping

Existing knowledge graph representation learning models do not utilize the semantic information of philosophical entities and philosophical relationships sufficiently, which makes it difficult for the models to obtain optimal representation performance. Therefore, this paper constructs a knowledge graph-based philosophical knowledge base and a text-enhanced knowledge graph representation learning model based on semantic propagation. Firstly, the ontology representation of the Chinese philosophy discipline was completed through six steps, and the ontology was generated by loop iteration. The modeling of the Chinese philosophy knowledge system was finished after the completion of semantic annotation. Then, the knowledge was stored and the ontology-based knowledge base of the Chinese philosophy discipline was created. The constructed philosophical knowledge base is used to identify entity texts and connect them with corresponding entities to create an enhanced knowledge map of Chinese philosophy-related texts. Attempting to use a graph convolutional network to propagate semantic information among nodes of the enhanced knowledge graph, the semantic information of text semantics and the ternary group is deeply fused, and the construction of the knowledge graph representation learning model is completed. Finally, experiments were conducted on the WN18RR dataset and the Chinese philosophy knowledge base constructed in this paper. The MRR value of this model on the Chinese philosophy knowledge base constructed in this paper is 0.046 more than that on the WN18RR dataset, and the majority of the “agree” and “strongly agree” on all indicators, with only one exception, indicating the validity and sophistication of the philosophical knowledge base and model constructed in this paper, which is conducive to the deepening of the cognitive logic of Chinese philosophy.