大模型增强下知识库语义扩展智能识别方法  

Intelligent Recognition Method for Semantic Extension of Knowledge Base Under Large Model Enhancement

作　　者：何剑萍 徐胜超 贺敏伟 HE Jianping;XU Shengchao;HE Minwei(School of Marxist,Guangzhou Huashang College,Guangzhou 511300;School of Artificial Intelligence,Guangzhou Huashang College,Guangzhou 511300)

机构地区：[1]广州华商学院马克思主义学院,广州511300 [2]广州华商学院人工智能学院,广州511300

出　　处：《计算机与数字工程》2025年第3期741-746,共6页Computer & Digital Engineering

基　　金：国家自然科学基金面上项目(编号:61972444);广州华商学院校内科研导师制项目(编号:2023HSDS26)资助。

摘　　要：为提升智能系统的处理能力和准确性,提出大模型增强下知识库语义扩展智能识别方法。将连续的文本切分成一系列独立的基本词汇,计算词汇的相似度,选出其中的关键词。针对关键词,利用本体扩展算法进行语义扩展。利用自注意力机制的深度学习方法,通过捕捉序列数据中的关键信息和上下文信息,实现知识库语义扩展智能识别。实验结果表明:所研究方法应用下,交并比相对更高,表明其识别标签与真实标签之间的重叠程度更高,说明该方法在识别准确性方面表现更好。In order to improve the processing capacity and accuracy of the intelligent system,the intelligent recognition method of expanding the knowledge base with large model enhancement is proposed.It divides the continuous text into a series of independent basic words,calculates the similarity of the words,and selects the key words.For the keywords,the ontology extension algorithm is used for semantic expansion.Deep learning approach using self-attention mechanism enables knowledge base semantic extended intelligent recognition by capturing key and contextual information in sequential data.The results show that under the application of the research method,the intersection ratio is relatively higher,indicating that the overlap between the identification label and the real label is higher,indicating that the method performs better in the identification accuracy.

关 键 词：大模型增强 分词 关键词提取 知识库语义扩展 智能识别 

分 类 号：TP31.84[自动化与计算机技术—计算机软件与理论]