深度学习中文本相似度计算研究综述  

REVIEW OF TEXTUAL SIMILARITY CALCULATION IN DEEP LEARNING

作　　者：徐传丽 周世杰[1] 吴春江 Xu Chuanli;Zhou Shijie;Wu Chunjiang(School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610054,Sichuan,China)

机构地区：[1]电子科技大学信息与软件工程学院,四川成都610054

出　　处：《计算机应用与软件》2024年第11期1-14,共14页Computer Applications and Software

基　　金：四川省科技厅重大专项(2019YFG0399,2020YFG0460);四川省科技厅重点研发项目(2019YFG0505)。

摘　　要：文本相似度计算是自然语言处理的重要任务之一,通过总结分析学术界提出的经典方法和最新进展,对深度学习中文本相似度计算的文本表示和相似度计算两个模块进行分类整理和全面综述。社交网络的发展衍生出了短文本相似度计算这一重要子任务,因此针对每一模块,梳理文本相似度计算相关技术与理论基础,总结其在短文本中的具体应用及改进;整理文本相似度计算领域的常用数据集及评价指标;讨论文本相似度计算未来可能的发展方向。Textual similarity calculation is one of the most important tasks of natural language processing.By summing up and analyzing the classical methods and the latest developments,the text representation and textual similarity calculation in deep learning were summarized and reviewed comprehensively.Due to the development of social networks,short textual similarity calculation,an important subtask has been derived.For each module,this paper sorted out the related technologies and theoretical foundations of textual similarity calculation,and summarized their specific applications and improvements in short texts.This paper summed up the common data sets and evaluation indexes in the field of textual similarity calculation.Some possible directions of textual similarity calculation were pointed out.

关 键 词：文本表示 文本相似度 短文本 自然语言处理 神经网络 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]