Querying structured and unstructured data utilizing hybrid natural language queries with semantic filtering

Abstract

The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing a hybrid query to generate a summary of requested data from a composite database. The disclosed systems receive a prompt requesting data from a composite database including structured data sources and unstructured data sources. The disclosed systems generate a hybrid query comprising one or more structured subqueries and one or more unstructured semantic subqueries. The disclosed systems determine an initial subset of data by executing the structured subqueries on the structured data sources. The disclosed systems determine an actionable subset of data from the initial subset of data by utilizing a set of semantic filters to analyze unstructured data sources of the initial subset of data according to the unstructured semantic subqueries. The disclosed systems utilize a large language model to analyze the actionable subset of data to generate a summary of the requested data for display.