Natural Language Interfaces for Tabular Data Querying and Visualization: A Survey

The emergence of natural language processing has revolutionized the way usersinteract with tabular data, enabling a shift from traditional query languagesand manual plotting to more intuitive, language-based interfaces. The rise oflarge language models (LLMs) such as ChatGPT and its successors has furtheradvanced this field, opening new avenues for natural language processingtechniques. This survey presents a comprehensive overview of natural languageinterfaces for tabular data querying and visualization, which allow users tointeract with data using natural language queries. We introduce the fundamentalconcepts and techniques underlying these interfaces with a particular emphasison semantic parsing, the key technology facilitating the translation fromnatural language to SQL queries or data visualization commands. We then delveinto the recent advancements in Text-to-SQL and Text-to-Vis problems from theperspectives of datasets, methodologies, metrics, and system designs. Thisincludes a deep dive into the influence of LLMs, highlighting their strengths,limitations, and potential for future improvements. Through this survey, we aimto provide a roadmap for researchers and practitioners interested in developingand applying natural language interfaces for data interaction in the era oflarge language models.