arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2308.15239cs.AI
收藏

低代码平台中的自然语言到SQL转换

Natural language to SQL in low-code platforms

Tripadvisor(猫途鹰);Zendesk(赞达思克);OutSystems;RedLight

Sofia Aparicio, Samuel Arcadinho, João Nadkarni, David Aparício, João Lages, Mariana Lourenço, Bartłomiej Matejczyk, Filipe Assunção

AI总结:

针对低代码平台中SQL查询编写难题,提出利用众包数据训练NL-to-SQL模型,并通过反馈循环持续优化,实现采用率提升240%和失败率降低90%。

摘要:

在低代码平台中,开发者面临的最大挑战之一是如何使用SQL查询从数据库中检索数据。在此,我们提出了一种流水线,允许开发者使用自然语言(NL)来检索数据。在本研究中,我们收集、标注并验证了覆盖OutSystems用户最常执行的SQL查询的数据。我们利用这些数据训练了一个生成SQL的自然语言模型。此外,我们描述了整个流水线,其中包括一个反馈循环,使我们能够快速收集生产数据并将其用于重新训练我们的SQL生成模型。通过众包方式,我们收集了26k个自然语言和SQL配对,并从生产数据中额外获得了1k个配对。最后,我们开发了一个用户界面,允许开发者在提示中输入自然语言查询,并接收生成的SQL查询的用户友好表示。我们使用A/B测试在生产环境中比较了四种不同的模型,观察到与首个投入生产的模型相比,该功能的采用率提高了240%,参与率提高了220%,失败率降低了90%,展示了我们的流水线在持续改进功能方面的有效性。

英文摘要:

One of the developers' biggest challenges in low-code platforms is retrieving data from a database using SQL queries. Here, we propose a pipeline allowing developers to write natural language (NL) to retrieve data. In this study, we collect, label, and validate data covering the SQL queries most often performed by OutSystems users. We use that data to train a NL model that generates SQL. Alongside this, we describe the entire pipeline, which comprises a feedback loop that allows us to quickly collect production data and use it to retrain our SQL generation model. Using crowd-sourcing, we collect 26k NL and SQL pairs and obtain an additional 1k pairs from production data. Finally, we develop a UI that allows developers to input a NL query in a prompt and receive a user-friendly representation of the resulting SQL query. We use A/B testing to compare four different models in production and observe a 240% improvement in terms of adoption of the feature, 220% in terms of engagement rate, and a 90% decrease in failure rate when compared against the first model that we put into production, showcasing the effectiveness of our pipeline in continuously improving our feature.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑