摘要
为构建一个面向幕墙企业和用户的产品营销和运维服务平台,本文设计并实现了面向幕墙的知识问答功能模块,此模块的主要内容包括基于爬虫的数据收集模块、用于知识库的构建与维护、敏感词过滤和基于Jieba分词等自然语言处理技术实现的数据预处理模块、基于Doc2Bow构建稀疏向量、文本相似度计算模块、知识问答功能实现模块。从而可以给用户提供问题的答案,比较迅速的满足用户的查询需求,节省用户在网页中寻找准确答案的时间,提高寻求答案的效率。
In this article an IQAS(intelligent question answering system)module oriented to curtain wall is designed and realized for the purpose of building a platform for marketing,operating and maintaining which is helpful for curtain wall enterprises and their users.This module is made up of 4 sections,which are data collection based on crawler,data pre-processing,sparse vector based on Doc2 Bow,calculation of text similarity and function realization of IQAS.It can provide the answers with the users and quickly meet their query requirements.It helps to save their time to search for the correct answers online and improve efficiency.
作者
常兵兵
罗才喜
张喆
CHANG Bingbing;LUO Caixi;ZHNAG Zhe(Tianjin Feiyu Curtain Wall Decoration Engineering Co.,Ltd.,Tianjin 300300;Binhai College,Nankai University,Tianjin 300270)
出处
《软件》
2021年第2期64-66,共3页
Software
基金
天津市智能制造专项资金项目(201810602)
中国高校产学研创新基金-新一代信息技术创新项目(2019ITA02028,2019ITA03011)
2021年度天津市交通运输科技发展项目(2021-19)。
关键词
运维服务
知识问答
稀疏向量
文本相似度
operation and maintenance
IQAS
sparse vector
text similarity