Abstract
NL2SQL (NLP Language To SQL) is a cutting-edge research direction of natural language processing, which converts natural query statements input by users into executable SQL statements. CCKS2022 proposes a multi-database multi-table NL2SQL task for the financial domain. For this task, this paper proposes a SQL generation method based on semantic parsing. The method adopts the multi-stage iterative generation mode of “question-database name-table name-column name-SQL statement”, uses semantic parsing and semantic similarity learning methods in acquiring table names, and generates and selects SQL statements based on the same query Multi-input statement for integrated filtering. At the end of the competition, the label replacement method was adopted, that is, the tables and columns in the SQL statement were replaced with tags to reduce the difficulty of generation.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Herzig∗, J., Shaw, P., Chang, M.-W., Guu, K., Pasupat, P., Zhang, Y.: Unlocking Compositional Generalization in Pre-trained Models Using Intermediate Representations (2021)
Andreas, J.: Good-enough compositional data augmentation (2020)
Scholak, T., Schucher, N., Bahdanau, D.: PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models (2021)
Mike, L.: BART: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (2021)
Gao, T., Yao, X., Chen, D.: SimCSE: Simple Contrastive Learning of Sentence Embeddings (2021)
Xue∗, L., et al.: mT5: A massively multilingual pre-trained text-to-text transformer (2020)
Li, D., et al.: Unified Language Model Pre-training for Natural Language Understanding and Generation (2019)
Choi, D., Shin, M.C., Kim, E., Shin, D.R.: RYANSQL: Recursively Applying Sketch-based Slot Fillings for Complex Text-to-SQL in Cross-Domain Databases (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Yi, X., Wang, R., Zhang, H., Zhen, S. (2022). Structured Design Solves Multiple Tables of NL2SQL. In: Zhang, N., Wang, M., Wu, T., Hu, W., Deng, S. (eds) CCKS 2022 - Evaluation Track. CCKS 2022. Communications in Computer and Information Science, vol 1711. Springer, Singapore. https://doi.org/10.1007/978-981-19-8300-9_24
Download citation
DOI: https://doi.org/10.1007/978-981-19-8300-9_24
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-8299-6
Online ISBN: 978-981-19-8300-9
eBook Packages: Computer ScienceComputer Science (R0)