Enterprise Data Graph Search via Link Analysis and Metadata Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing search techniques for normalized data are inefficient in returning highly relevant results for unstructured queries, as they often produce irrelevant results or require schema awareness and significant computational resources.
Innovation Solution
The system identifies highly relevant search results by generating a graph of normalized data objects and metadata, using link analysis and metadata scoring to prioritize candidate nodes, and locating search result subgraphs that match multi-term queries, allowing for early termination of the search process to return the most relevant results quickly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing search techniques are used for normalized data, then the search process can be completed, but the results are often irrelevant and computational resources are excessive
Solution Approach 1:
The search process is segmented into multiple phases: candidate node identification, priority scoring, and subgraph search. This segmentation allows the system to process only relevant portions of the graph rather than examining the entire normalized data structure, thereby improving both result relevance and computational efficiency.
Solution Approach 2:
The system performs preliminary actions by pre-computing metadata scores and link analysis for candidate nodes before executing the actual search. This preliminary processing enables early termination of the search process when highly relevant results are found, avoiding unnecessary computational resources while maintaining high result relevance.
2Measurement precision
If schema awareness is required for searching normalized data, then search accuracy improves, but ease of operation deteriorates
Solution Approach 1:
The patent introduces metadata as an intermediary layer between the user's unstructured query and the normalized data schema. The metadata layer automatically maps user-friendly search terms to relevant data objects without requiring users to understand the underlying schema, thereby maintaining search accuracy while improving ease of operation.
Solution Approach 2:
The system replaces the mechanical requirement for schema-aware queries with an automated metadata-driven search mechanism. Instead of requiring users to manually specify schema constraints, the system uses metadata scoring and link analysis to automatically identify relevant results, substituting user effort with automated intelligent processing.
3Reliability
If the entire graph is examined to ensure complete search results, then result completeness improves, but search time increases
Solution Approach 1:
The system applies partial action by examining only the necessary portions of the graph required to find highly relevant results. Through priority-based candidate node selection and early termination strategies, the system achieves sufficient result completeness without the need to examine the entire graph, thereby reducing search time while maintaining reliability.
Solution Approach 2:
The search process incorporates feedback mechanisms where metadata scores and link analysis results continuously refine the search strategy. This feedback allows the system to dynamically adjust the search scope and terminate early when highly relevant results are found, balancing result completeness with reduced search time.
Data Source
AI summary
Highly relevant search results for unstructured queries are efficiently identified from normalized data. A server graphs relationships between each of the data objects. For each search term, the server identifies a candidate set of data objects mapped to the term. The server calculates priority scores for the candidate data objects based at least in part on one or more of: a link analysis of the graph; or an analysis of metadata describing structural constraints upon the candidate data objects. Based on the graph, the server identifies one or more search result subgraphs, each comprising at least one data object from each of the candidate sets. The server looks for subgraphs in an order that is based on the priority scores for the candidate data objects. The server may terminate its search early, in part because prioritizing the candidate data objects increases the likelihood of receiving relevant search results first.