Context-Based Entity Name Spell Correction for Search Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search engines struggle to correct misspelled entity names, especially those of ordinary entities, as they often do not appear in dictionaries, leading to difficulty in identifying relevant search results.
Innovation Solution
A system that derives name-context pairs from various text sources, calculates context consistency measures, and stores context-entity name data to correct misspelled entity names by leveraging additional context terms, using popularity scores and similarity analysis to suggest accurate alternatives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional dictionary-based spell correction is used, then common words can be corrected, but entity names (especially ordinary entities) cannot be corrected because they do not appear in dictionaries
Solution Approach 1:
The patent transitions from one-dimensional dictionary lookup to multi-dimensional context analysis by examining surrounding terms, document titles, and co-occurring entities to infer correct spellings of entity names that are not present in traditional dictionaries
Solution Approach 2:
The patent introduces context terms as intermediaries between the misspelled entity name and its correct form, using surrounding textual context to bridge the gap when direct dictionary matching fails
2Adaptability or versatility
If context-based correction is implemented, then entity names can be corrected, but system complexity increases due to processing multiple text sources and calculating consistency measures
Solution Approach 1:
The patent performs preliminary actions by pre-processing text sources to extract and store name-context pairs with their consistency measures in advance, so that during query processing, the system can quickly retrieve and apply pre-computed context information without performing complex calculations in real-time
Solution Approach 2:
The patent segments the correction process into distinct phases: text collection, name-context pair extraction, consistency measure calculation, and query-time application, allowing each component to be optimized independently and processed in manageable stages
3Measurement precision
If multiple text sources are processed to improve correction accuracy, then more context information is available, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-processing and analyzing multiple text sources offline to build a comprehensive context database with pre-computed consistency measures, transforming the expensive multi-source analysis into a one-time cost that enables fast online query processing
Solution Approach 2:
The patent applies partial action by selectively processing and weighting different text sources based on their relevance and quality, focusing computational resources on the most informative sources rather than uniformly processing all available text
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for correcting entity names. One method includes receiving texts and deriving a plurality of name-context pairs from the texts. The method further includes calculating a context consistency measure for each name-context pair and storing context-entity name data representing the name-context pairs. Another method includes identifying an entity name and one or more context terms from a query and generating candidate names for the entity name. The method further includes determining a score for each of the candidate names, selecting a number of top scoring candidate names, and using the selected candidate names to respond to the query.