首页|基于认知技能图谱的智能问答系统设计与实现

基于认知技能图谱的智能问答系统设计与实现

扫码查看
针对电气专业学生获取知识烦琐、实践能力提升困难的问题,该文提出并开发基于认知技能图谱的电力电子开关电源实验智能问答系统,旨在为学生提供个性化指导.首先,引入贝叶斯心理测量模型评估学生能力达成情况;然后,利用知识图谱技术构建陈述性知识图谱和用户数字镜像,分别用于存储电源特定领域知识和用户能力信息;最后,基于问句分析、模板匹配、答案生成3个模块开发电源实验智能问答系统.该系统利用HanLP对学生输入问题进行分词处理和词性标注,使用朴素贝叶斯分类器进行问题分类,并与相应的问题模板进行匹配.根据模板和关键词,系统生成相应的Cypher查询语句,结合学生的能力信息,在知识图谱中搜索符合条件的答案.实验结果表明,该系统能够有效地识别电气专业学生的能力缺陷,并提供个性化的问答服务,实现了更高效、精准的智能辅助.
Design and implementation of intelligent question answering system based on cognitive skill atlas
In order to solve the problem that electrical engineering students have cumbersome knowledge ac-quisition and difficulty in improving practical ability,this paper proposes and develops an intelligent question and answer system for power electronic switching power supply experiment based on cognitive skill graph,aiming to provide personalized guidance for students.Firstly,the Bayesian psychometric model was intro-duced to evaluate the achievement of students'abilities,and then the knowledge graph technology was used to construct a declarative knowledge graph and a user digital image,which were used to store the knowledge of specific fields of power supply and user ability information,respectively.The system uses HanLP to per-form word segmentation and part-of-speech annotation of students'input questions,and uses a naive Bayes classifier to classify the questions,which are matched with the corresponding question templates.According to the template and keywords,the system generates the corresponding Cypher query statement,and sear-ches for eligible answers in the knowledge graph based on the student's ability information.Experimental re-sults show that the system can effectively identify the ability defects of electrical students,provide personal-ized Q&A services,and achieve more efficient and accurate intelligent assistance.

Bayesian psychometric modelcognitive skills atlasknowledge graphquestion answering system

蔡令仪、段斌、旷怡、柯其聪

展开 >

湘潭大学 自动化与电子信息学院,湖南湘潭 411105

贝叶斯心理测量模型 认知技能图谱 知识图谱 问答系统

湖南省自然科学基金湖南省新工科研究与实践项目

2020JJ6034202012

2024

湘潭大学学报(自然科学版)
湘潭大学

湘潭大学学报(自然科学版)

CSTPCD
影响因子:0.403
ISSN:2096-644X
年,卷(期):2024.46(2)
  • 12