基于NL2SQL的智能搜索系统研究  

Research on Intelligent Search System based on NL2SQL

作　　者：展昭 李双远[2] ZHAN Zhao;LI Shuangyuan(School of Information and Control Engineering,Jilin Institute of Chemical Technology,Jilin City 132022,China;Information Construction Office,Jilin Institute of Chemical Technology,Jilin City 132022,China)

机构地区：[1]吉林化工学院信息与控制工程学院,吉林吉林132022 [2]吉林化工学院信息化建设办公室,吉林吉林132022

出　　处：《吉林化工学院学报》2023年第7期65-70,76,共7页Journal of Jilin Institute of Chemical Technology

摘　　要：在当今大数据时代,信息繁多,但是真正对自己有价值的数据很少,如何从冗杂的数据里面检索出自己真正需要的数据至关重要。传统的搜索系统只能够对已经出现的数据进行展示,不能够对数据进行整合、计算等处理。因此使用NL2SQL技术来实现搜索系统,首先构建异构图结构对数据进行预处理,然后通过相对位置的自注意力机制将异构图编码成高维隐层向量,最后通过树形解码器将高维隐层向量解码成SQL语句。这样能够更好地满足用户真实需求,用户能够更加自由地和数据库进行交互,充分挖掘出数据的价值。In the era of big data,there is a lot of information,but there is very little data that is really valuable to you.It is crucial to retrieve the data you really need from the redundant data.The traditional search system is only able to display the data that has appeared,not able to integrate,calculate and other processing of the data.This paper used NL2SQL technology to implement the search system.First,we constructed the heterogeneous graph structure to preprocess the data,then encoded the heterogeneous graph into a high-dimensional hidden layer vector by the self-attention mechanism of relative position,and finally decoded the high-dimensional hidden layer vector into SQL statements by the tree decoder.This can better meet the real needs of users,who can interact with the database more freely and fully exploit the value of the data.

关 键 词：NL2SQL 搜索系统 异构图 数据库 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]