Research on Modeling the Semantic Relationship of English Denominal Verbs Based on Knowledge Graphs and Extracting Translation Corresponding Rules - Taking English-Chinese Translation as an Example

Abstract

Denominal verbs are verbs converted from concrete nouns through zero conversion, and their semantic relationships are complex, making English-Chinese translation highly uneven. Traditional translation methods often affect translation quality, while current semantic relationship modeling of English denominal verbs lacks systematicity and does not fully exploit knowledge-graph advantages. Manual extraction of translation correspondence rules has low universality and weak adaptability, which is insufficient for intelligent translation systems and technical English translation in engineering fields such as electromagnetic waves, antennas, and propagation. This paper first summarizes the semantic features of English denominal verbs and the status of English-Chinese translation. It then constructs a knowledge-graph-based semantic relationship modeling framework, clarifies node associations, and designs semantic relationship extraction and rule extraction algorithms. Finally, multi-domain corpora are used to verify the effectiveness of the model and rules. Experimental results show that semantic recognition accuracy reaches 89.6%, and the adaptation rate of extracted translation correspondence rules reaches 87.3%, both significantly outperforming traditional methods and improving translation accuracy and consistency.