Abstract
Systems that translate natural language to database queries may encounter challenges with ambiguous terms, for example, 'high-value' or 'recent'. A system for interactive visual disambiguation can visually identify and highlight uncertain terms within a user’s natural language input. A user can interact with a highlighted term to generate a manipulatable graphical tether that connects the term to an interactive logic card. This card may provide controls, such as sliders, for refining the definition of the term. As a user adjusts a definition, a data projection engine can update a data preview in near real-time to illustrate the potential impact of the change. This process can create a more transparent workflow for aligning user intent with a database schema prior to query execution.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Hoyer, Timo, "Interactive Visual Disambiguation for Natural Language Database Queries", Technical Disclosure Commons, (March 11, 2026)
https://www.tdcommons.org/dpubs_series/9496