Construction of a Semantic Network for International Chinese Language Education Based on Knowledge Graph Technology and Optimization of Its Teaching Resources

In the era of big data, technology is changing rapidly, Chinese education resources supported by the Internet are growing exponentially, and many problems such as confusing conceptual elaboration and unclear relationship with everyday language can occur in the process of international Chinese education promotion and textbook use. Therefore, based on the advantage that knowledge graph can describe the relationship between words or statements concisely, efficiently and quickly, this paper proposes a knowledge enhancement and cue tuning embedding model based on multimodal knowledge graph embedding technology by utilizing text and image extraction to supplement the knowledge graph based on the graphic needs of Chinese education teaching. On this basis, a knowledge semantic network for Chinese education is constructed. Through experimental analysis, this paper after knowledge enhancement for text entity extraction task effect enhancement help more. This paper’s model got the best results for each metric compared to other benchmark models, and in large sample scenarios, this paper’s model is better, with precision, recall, and F1 values of 87.42%, 88.03, and 87.23 on the CTec2020 dataset, respectively. Meanwhile, in the cross-task scenario test, the F1 value of this paper’s model is 78.8 and 76.61 on CTec2018 and CTec2020, respectively, with the optimal results for each evaluation index, which further proves the performance of this paper’s model. The number of entities extracted from the semantic network of this paper for various words in Chinese education is more than 3,000, the average word length of the five corpus is 2.994, and the accuracy rate of entity extraction ranges from 91.05% to 97.66%, with the highest points of 98, 221, 132, 159, and 147, respectively, which is a great advantage in all aspects, showing that the semantic network constructed for Chinese education in this paper has a great advantage in accuracy, comprehensiveness, and domain. It shows that the semantic network constructed in this paper for Chinese language education has improved in accuracy, comprehensiveness and domain, and the quality of entities is better. This research knowledge semantic network can be practically applied in the field of Chinese education, which provides scientific support for the optimization of Chinese educational resources.