摘要
近年来,随着人工智能技术的日益发展,基于自然语言理解的信息查询系统也逐渐成为现实。针对基于关键词搜索的搜索引擎难以精准捕获用户搜索意图的现状,提出了一种基于BERT的意图识别和语义槽填充联合模型,将其应用于搜索系统可以支持用户输入非结构化查询语句进行搜索。实验表明该模型能够识别用户的搜索意图从而提高搜索的准确率和速度。
In recent years,with the increasing development of artificial intelligence technology,make the information query system based natural language understanding possible gradually.Since it is difficult for the user to capture the search intention accurately us⁃ing search engine based on keywords,this paper proposes a joint intent classification and slot filling model based on BERT which used to the search engine can support the user to input the unstructured sentences.Experimental results demonstrate that our proposed mod⁃el can identify the user's search intent so as to improve the accuracy and speed of the search.
作者
闫峥
杨砾
江强
任艾
张沛然
许青青
Yan Zheng;Yang Li;Jiang Qiang;Ren Ai;Zhang Peiran;Xu Qingqing(Science and Technology Department for Shanghai Public Security Bureau,Shanghai 200042;Shanghai Jiao Tong University,Shanghai 200240;DATATOM,Shanghai 200030)
出处
《现代计算机》
2021年第23期48-52,共5页
Modern Computer
基金
上海市科学技术委员会科研计划项目“:城市大脑”一期“/平安大脑”关键技术及应用研究(18DZ1200900)。
关键词
意图识别
槽填充
语义搜索
BERT
Intent Classification
Slot Filling
Semantic Search
BERT
