Scholarlink

<p>Interactive teaching evaluation plays an important role in the development of modern education. Accurately evaluating teaching effectiveness can effectively promote the improvement of classroom education. However, traditional methods of interactive teaching evaluation face limitations such as low evaluation efficiency and delayed feedback. Therefore, this study raises an interactive teaching evaluation model based on Knowledge Graph and Complex Question Answering. The model uses K-dimensional Tree and Sentence Bidirectional Encoder Representations from Transformers variant to optimize the retrieval process, while introducing Graph Convolution Neural Networks and attention mechanism to aggregate entity neighbor feature information. In addition, the model incorporates Hybrid Particle Swarm Optimization and Bidirectional Long Short-Term Memory networks to construct a complete interactive teaching evaluation model. Experimental results show that the improved algorithms achieve high fitness and classification accuracy of 95.11%. At the same time, the classification accuracy of the interactive teaching evaluation model built in the simulation verification reached 96.82%, In the K12 education stage, question–answer matching and answer accuracy reach 89.36% and 97.74%, respectively, while on the dialogue-based question answering dataset, they reach 87.38% and 98.47%. In the practical application verification, the activity level of the model classroom is highly consistent with the evaluation of the expert group, the error of the number of questions is only 2, 1, 2 times, the convergence time of the subject scene switching is 0.96&#xa0;s, and the accuracy rate of scene recognition is 95.67%. These results indicate that the Knowledge Graph and Complex Question Answering based interactive teaching evaluation model can improve question–answer recognition efficiency and evaluation accuracy. This study contributes to the precise construction of interactive teaching evaluation models in educational scenarios and promotes the steady development of modern education.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interactive Teaching Evaluation Method Based on KG-CQA

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月17日

查看全文
  • Shaojiao Qin,
  • Lin Li

摘要

Interactive teaching evaluation plays an important role in the development of modern education. Accurately evaluating teaching effectiveness can effectively promote the improvement of classroom education. However, traditional methods of interactive teaching evaluation face limitations such as low evaluation efficiency and delayed feedback. Therefore, this study raises an interactive teaching evaluation model based on Knowledge Graph and Complex Question Answering. The model uses K-dimensional Tree and Sentence Bidirectional Encoder Representations from Transformers variant to optimize the retrieval process, while introducing Graph Convolution Neural Networks and attention mechanism to aggregate entity neighbor feature information. In addition, the model incorporates Hybrid Particle Swarm Optimization and Bidirectional Long Short-Term Memory networks to construct a complete interactive teaching evaluation model. Experimental results show that the improved algorithms achieve high fitness and classification accuracy of 95.11%. At the same time, the classification accuracy of the interactive teaching evaluation model built in the simulation verification reached 96.82%, In the K12 education stage, question–answer matching and answer accuracy reach 89.36% and 97.74%, respectively, while on the dialogue-based question answering dataset, they reach 87.38% and 98.47%. In the practical application verification, the activity level of the model classroom is highly consistent with the evaluation of the expert group, the error of the number of questions is only 2, 1, 2 times, the convergence time of the subject scene switching is 0.96 s, and the accuracy rate of scene recognition is 95.67%. These results indicate that the Knowledge Graph and Complex Question Answering based interactive teaching evaluation model can improve question–answer recognition efficiency and evaluation accuracy. This study contributes to the precise construction of interactive teaching evaluation models in educational scenarios and promotes the steady development of modern education.