摘要
在科研项目申报过程中,当前多采用人工方式进行评审专家遴选,由于人工对领域知识的理解有限,且具有一定的主观性倾向,随着项目申报数量的增加,人工选择的效率和准确率逐渐降低.为解决此问题,该文提出一种基于文本相似度的评审专家推荐方法.通过对项目论文信息进行数据挖掘,利用编辑距离模糊匹配和Wordnet语义扩展方法改进文本相似度计算,设计对比实验分别说明方法的可行性,并对推荐结果给出解释.实验结果表明,该文方法能够有效解决评审专家遴选问题.
In the process of applying for scientific research projects,the selection of review experts is often carried out manually.Due to the limited understanding of domain knowledge and the subjective tendency of manual selection,the efficiency and accuracy of manual selection gradually decrease with the increase of the number of project declarations.To solve this problem,this paper proposes a method of expert recommendation based on text similarity.Through data mining of project paper information,the text similarity calculation is improved by usingediting distance fuzzy matching and Wordnet semantic extension methods.The validity of the method is illustrated by designing comparative experiments,and the recommendation results are explained.The experimental results show that this method can effectively solve the problem of selecting evaluation experts.
出处
《科技资讯》
2019年第17期173-176,共4页
Science & Technology Information
关键词
专家推荐
数据挖掘
文本相似度
语义扩展
Expert recommendation
Data mining
Text similarity
Semantic extension
