Schema-Free In-Graph Indexing for Knowledge Graph Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graph databases struggle with efficient analytics operations involving constraints on attributes associated with traversal in knowledge graphs, and external indexing engines are not effectively usable with these graphs.
Innovation Solution
Implementing schema-free in-graph indexing within the graph database system, creating multiple indexes that can be correlated to generate multi-dimensional indexes, facilitating query processing through index-to-index traversals and optimizing query optimization techniques without relying on external indexing engines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If existing graph databases are tuned for relationship exploration queries, then query traversal performance is improved, but analytics operations with attribute constraints become inefficient
Solution Approach 1:
The patent segments the indexing function by creating multiple specialized single-dimension indexes (e.g., vertex property indexes, edge property indexes, path indexes) that can be independently optimized for different query types. This allows the system to maintain fast traversal performance while enabling efficient analytics operations by selecting appropriate indexes based on query requirements.
Solution Approach 2:
The patent introduces multi-dimensional indexes that correlate multiple single-dimension indexes across different dimensions (vertex properties, edge properties, path attributes). This dimensional expansion enables the system to handle complex analytics operations with multiple attribute constraints while maintaining the efficiency benefits of specialized single-dimension indexes.
2Adaptability or versatility
If external indexing engines are used for graph data, then indexing capability is provided, but integration with knowledge graphs is ineffective
Solution Approach 1:
The patent merges the indexing engine directly into the graph database system, creating a unified architecture where indexes are created, stored, and managed within the same system that stores the graph data. This integration eliminates the complexity of external engine connections and enables seamless coordination between graph traversal and index-based analytics operations.
Solution Approach 2:
The patent creates a universal indexing framework that handles multiple data types and query patterns through a single integrated system. The indexing engine supports vertex properties, edge properties, path attributes, and various query types (traversal, analytics, constraints) through a unified interface, eliminating the need for separate external indexing solutions.
3Adaptability or versatility
If multiple single-dimension indexes are created and correlated, then multi-dimensional indexing capability is improved, but system complexity increases
Solution Approach 1:
The patent performs preliminary actions by automatically creating and correlating single-dimension indexes when the graph database is initialized or when data is inserted. The system pre-computes and stores index structures in optimized formats, so that when multi-dimensional queries are executed, the correlation of indexes occurs efficiently without ad-hoc complexity.
Solution Approach 2:
The patent introduces an intermediary indexing layer that sits between the raw graph data and the query processing engine. This intermediary layer manages the correlation of multiple single-dimension indexes through standardized interfaces and data structures, shielding the complexity of multi-dimensional index management from both the data storage and query execution components.
Data Source
AI summary
Methods, systems, and computer program products for schema-free in-graph indexing are provided herein. A computer-implemented method includes creating multiple indexes directed to data within a knowledge graph; correlating two or more of the created indexes, thereby generating one or more multi-dimensional indexes; determining, based on a received query, one or more traversal paths within the data of the knowledge graph and the generated multi-dimensional indexes, wherein the traversal paths facilitate processing of the query; and outputting a response to the query based on the determined traversal paths.