Space Of Interest Database Search Using Hierarchical Relationship Filtering

The patent divides the search space into hierarchical levels: region objects (e.g., districts, cities) containing multiple space objects (e.g., landmarks, buildings), which in turn contain multiple POIs. This segmentation allows the system to first filter by region, then by space objects within that region, significantly improving search accuracy while maintaining manageable system complexity through organized data structures.

The patent introduces a spatial containment dimension by defining region objects that encompass multiple space objects, and space objects that contain multiple POIs. This creates a hierarchical containment structure where POIs are not just points on a map but are nested within space objects which are nested within regions, adding a dimensional layer to the search space that enables precise area-based filtering.

The patent performs preliminary filtering by region and space object containment before executing the keyword search. The system first identifies the target region object and its contained space objects, then performs the POI search only within those pre-defined boundaries. This preliminary spatial filtering action ensures that keyword matches are restricted to the user's area of interest, preventing irrelevant results while maintaining search efficiency.

The patent introduces space objects as intermediary entities between region objects and POIs. These space objects act as mediators that provide spatial context and containment relationships, allowing the system to filter POIs by their location within specific space objects or regions while still performing keyword-based searches. This intermediary layer preserves search efficiency while eliminating irrelevant results.