计算机应用与软件2023，Vol.40Issue(12)：33-40,8.DOI:10.3969/j.issn.1000-386x.2023.12.006
基于多角度交叉注意力机制的知识库问答方法
QUESTION ANSWERING OVER KNOWLEDGE BASE BASED ON MULTI-VIEW CROSS-ATTENTION MECHANISM
摘要
Abstract
In recent years,question answering over knowledge base methods usually use multi-view information to represent candidate answers,ignoring the mutual influence between these information,and they calculate the correlation between the word of the question and the multi-view information of the answer,ignoring the overall and detailed information of the two.Based on the above problems,a multi-view cross-attention model is proposed.Multi-view cross-attention mechanism was used to obtain the cross-influence between multi-view information of the candidate answer.The question and the candidate answer information were represented as a whole,and the two-way cross-attention mechanism was used to calculate the correlation between the two at the overall level.The correct rate of obtaining the answer was improved.Experiments were conducted on the FreeBase knowledge base and WebQuestions data sets,and the F1 value reached 55.84%,which was better than the recent methods that performed well.关键词
知识库问答/多视角信息/多视角交叉注意力机制/双向交叉注意力机制Key words
Question answering over knowledge base/Multi-view information/Multi-view cross-attention mechanism/Two-way cross-attention mechanism分类
信息技术与安全科学引用本文复制引用
代发扬,符海东,高峰,顾进广..基于多角度交叉注意力机制的知识库问答方法[J].计算机应用与软件,2023,40(12):33-40,8.基金项目
国家自然科学基金项目(U1836118,61673304) （U1836118,61673304）
国家社科基金重大计划项目(11&ZD189) （11&ZD189）
湖北省自然科学基金项目(2018CFB194). （2018CFB194）
