BookSQL: A Large Scale Text-to-SQL Dataset for Accounting Domain
This addresses a gap for non-technical users in accounting, but it is incremental as it extends existing dataset efforts to a new domain.
The authors tackled the lack of large-scale Text-to-SQL datasets for finance and accounting by introducing BookSQL, a dataset with 100k query-SQL pairs and 1 million database records, and found significant performance gaps in existing models like GPT-4 when tested on it.
Several large-scale datasets (e.g., WikiSQL, Spider) for developing natural language interfaces to databases have recently been proposed. These datasets cover a wide breadth of domains but fall short on some essential domains, such as finance and accounting. Given that accounting databases are used worldwide, particularly by non-technical people, there is an imminent need to develop models that could help extract information from accounting databases via natural language queries. In this resource paper, we aim to fill this gap by proposing a new large-scale Text-to-SQL dataset for the accounting and financial domain: BookSQL. The dataset consists of 100k natural language queries-SQL pairs, and accounting databases of 1 million records. We experiment with and analyze existing state-of-the-art models (including GPT-4) for the Text-to-SQL task on BookSQL. We find significant performance gaps, thus pointing towards developing more focused models for this domain.