KathDB-FAO speeds up natural language queries in multimodal databases
KathDB-FAO: Synthesized Query Plans in a Multimodal DBMS
DatabasesArtificial Intelligence
Summary
Databases that handle multiple types of data can be hard to ask questions of using regular language. The authors created KathDB-FAO, a new system that turns natural language questions into efficient step-by-step plans for the database to follow. It does this by breaking down questions into small actions, linking them carefully, and creating special functions to run those actions quickly. Tests show that this approach can run queries much faster while keeping accuracy.
What this means in practice
- •For database administrators: Run natural language queries on databases with different data types more efficiently and accurately.
- •For software engineers: Build applications that process user questions in natural language into optimized database queries for faster results.
Authors
Guorui Xiao, Douglas Brown, Artur Borycki, Magdalena Balazinska
Abstract
We design, implement, and evaluate KathDB-FAO, a new query evaluation subsystem for our KathDB multimodal DBMS. KathDB-FAO takes as input a query in natural language (NL) and converts it into a query execution plan where each operator is a function whose body is synthesized during query evaluation, which allows powerful query-specific optimizations. To generate accurate and efficient plans from NL, KathDB-FAO first extracts fine-grained atomic actions for correctness, then establishes contracts on the inputs and outputs of those actions and groups them for efficiency, and finally synthesizes the function for each group on the fly. On SemBench, KathDB-FAO cuts execution cost by 58.8% on average across scenarios compared with the next best system, at comparable or better quality.