| 摘 要: 世界上很多高价值的数据信息储存在关系数据库中，访问这些数据需要掌握专门的结构化查询语言(SQL)，普通人很难直接使用。基于对现有对话机器人存在的问题和相关关键技术的梳理，本文融合了数据仓库、数据同步、数据库查询、消息推送、自然语言理解及语音识别等相关技术及产品，设计了数据库驱动的对话机器人。方案可以实现用户理解、消息推送、事实数据查询和分析数据查询四个功能，使得用户能够快速地获取信息。本文提出的数据库驱动的对话机器人具有较强的泛化性和可扩展性。 |
| 关键词: 关系型数据库 对话机器人 SQL 消息推送 |
|
中图分类号: TP183
文献标识码: A
|
|
| Research on Database-driven Chatbot |
|
LAI Guanjun1,2, YU Dan1,2, YAN Xiaoyu1,2, XIAO Peng1,2
|
( 1. Dalian Neusoft University of Information, Dalian 116023, China ;
2. Research Institute, Dalian Neusoft Education Technology Group Co. Limited, Dalian 116023, China)
laiguanjun@neuedu.com; yudan@neuedu.com; yanxiaoyu@neuedu.com; xiaopeng@neuedu.com
|
| Abstract: Many high-value data in the world are stored in relational databases. Access to these data requires a special Structured Query Language (SQL), which is difficult for ordinary people to use directly. Aiming at problems of the existing chatbot, this paper proposes to design database-driven chatbots by integrating related technologies and products, such as data warehouse, data synchronization, database query, message push, natural language understanding and speech recognition. The solution realizes four functions: user understanding, message push, fact-data query and analysis-data query, so that users can quickly obtain information. Database-driven chatbots proposed in this paper has a capability of strong generalization and scalability that enable chatbots to work with data-driven conversations. |
| Keywords: relational database chatbot SQL message push |
