Enhancing User-Centric Information Retrieval: A Unified Dual-DBMS Strategy for Integrating Full-Text and Knowledge Graph Searches

This article introduces a new search query language aimed at non-technical users that combines the simplicity of natural language queries with the precision of structured database queries. A distinctive feature of the language is the combination of the simplicity of well-known search languages, inherent in Google and Yandex, and advanced search capabilities for traversing knowledge graphs. We further detail the implementation of this language through a combination of two separate database management systems: Open Search for fulltext indexing and PostgreSQL for knowledge graph storage. In addition, we introduce a synchronization mechanism to maintain state consistency across these DBMS backends, ensuring data integrity and increasing the reliability of search results.