Spatial Index Structure for Navigation Destination Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Full text search systems using inverted indexes face performance issues when handling multiple search terms, leading to slower query response times and increased secondary storage consumption, especially when dealing with a large number of documents.
Innovation Solution
Implementing a spatial index structure, such as an R-tree, to model documents as low-dimensional vectors and store them, which reduces secondary storage usage and improves query times by clustering documents based on terms, particularly useful for structured low-dimensional documents like those in destination searches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If an inverted index is used to store document identifiers for each search term, then the system can efficiently retrieve documents containing specific terms, but the secondary storage consumption increases and query response time slows down when handling multiple search terms
Solution Approach 1:
The patent transforms the traditional inverted index from a flat key-value structure into a multi-dimensional spatial structure. Documents are represented as vectors in a k-dimensional space where each dimension corresponds to a search term. This dimensional transformation allows the system to leverage spatial indexing algorithms that can handle multiple search terms simultaneously, reducing the time complexity from linear iteration over document identifiers to efficient spatial range queries.
Solution Approach 2:
The system changes the parameter representation by converting document identifiers and search terms into numerical vectors with specific dimensions. Each document is represented as a vector where the position and value correspond to the presence and frequency of search terms. This parameter transformation enables the use of spatial distance metrics and clustering algorithms to improve query efficiency while maintaining search accuracy.
2Reliability
If an inverted index stores all document identifiers for each term, then complete search results can be obtained, but the secondary storage space is significantly consumed
Solution Approach 1:
The patent merges multiple inverted index structures into a unified spatial index. Instead of maintaining separate lists of document identifiers for each search term, the system combines all term-document relationships into a multi-dimensional vector space representation. This consolidation reduces redundant storage while preserving the ability to retrieve complete search results through spatial range queries that can efficiently filter documents based on multiple terms simultaneously.
Solution Approach 2:
The system creates a compressed vector representation of documents that captures essential searchability information without storing complete document identifier lists. Each document is represented as a compact vector in k-dimensional space, which serves as a compressed copy of the document's term relationships. This vector representation consumes significantly less storage space while maintaining the ability to reconstruct complete search results through spatial queries.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and system for full text search during destination selection using a navigation system is disclosed. The full text search system includes a relation table and a spatial index structure, e.g., an R-tree. The relation table maps tokens to a token identifier. Each level of a destination is mapped to its own dimension, e.g., Country to X, City to Y, and Street to Z. Each document is then mapped to an n-dimensional vector using the token identifiers.