Sub-query Evaluation for Multi-Representation Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data storage methods in computing systems often rely on canonical representation, which fails to account for the ambiguous nature of information, leading to incomplete or inaccurate data retrieval when multiple representations of a data item exist, especially in cognitive data graphs where different contexts require different representations of the same data.
Innovation Solution
The method involves determining and evaluating sub-queries for a query on a data graph that stores multiple representations of a data item, allowing for the selection of the most appropriate representation based on the query's context, using techniques such as natural language processing and numeric likelihood analysis to ensure accurate data retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional canonical representation methods are used, then data storage is simple, but data retrieval accuracy deteriorates when multiple representations exist
Solution Approach 1:
The patent segments the query evaluation process into multiple sub-queries, each corresponding to a different representation of the data item. Instead of evaluating a single complex query, the system divides the query into separate sub-queries that can be evaluated independently against different representations, improving accuracy while managing complexity through structured decomposition.
Solution Approach 2:
The patent changes the evaluation parameters by introducing numeric likelihood values associated with each representation. The system evaluates sub-queries using these likelihood parameters to determine the most appropriate representation, transforming the query evaluation from a binary match to a weighted assessment that accounts for multiple representations.
2Adaptability or versatility
If multiple representations of data items are stored, then adaptability improves, but query processing time increases
Solution Approach 1:
The patent performs preliminary action by pre-establishing the mapping between queries and sub-queries before actual query processing. The system pre-determines which sub-queries correspond to which representations, so that when a query arrives, the system can directly evaluate the relevant sub-queries without performing complex analysis, thus reducing processing time while maintaining adaptability.
Solution Approach 2:
The patent implements feedback mechanisms where the evaluation of sub-queries provides information about which representations are most appropriate. The system uses the results of sub-query evaluation to determine the final answer, creating a feedback loop that efficiently narrows down the most suitable representation based on actual query results rather than relying solely on pre-computed data.
3Measurement precision
If sub-queries are evaluated to select appropriate representations, then data retrieval accuracy improves, but computational resources increase
Solution Approach 1:
The patent applies partial action by evaluating only the necessary sub-queries rather than all possible representations. The system determines which sub-queries are relevant based on the query context and evaluates only those, avoiding unnecessary computational resources while maintaining high accuracy in representation selection by focusing on the most promising candidates.
Data Source
AI summary
A query is for retrieving a data item of a data graph. The data graph stores representations of the data item. Each representation stores information represented by the data item in a different manner or way. Sub-queries for the query each correspond to a different representation in which the data graph stores the data item. The sub-queries are evaluated to determine an appropriate representation of the data item in satisfaction or fulfillment of the query.