Geographic Entity Synonym Generation for Query Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of geographic information and mapping systems often enter queries with incorrect spellings of geographic entities, as they may use either compounded or decompounded forms, leading to failure in retrieving the intended information since the database stores only one correct form.
Innovation Solution
A computer-implemented method that identifies geographic entities with original names comprising multiple terms, forms name synonym rules by associating adjacent terms with single-word concatenations, validates these rules using a dictionary, and uses the single-word concatenations as synonyms to enhance query matching, allowing for correct retrieval even with incorrect spellings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the geographic database stores only the original compounded form of entity names, then data consistency and storage efficiency are maintained, but users cannot retrieve entities when using decompounded form queries
Solution Approach 1:
The system performs preliminary action by automatically generating synonym names (both compounded and decompounded forms) for geographic entities before users submit queries. This advance preparation ensures that regardless of whether users query with compounded or decompounded forms, the entity will be found in the database without requiring real-time analysis or manual intervention.
Solution Approach 2:
The patent introduces synonym names as an intermediary layer between the original entity names and user queries. These synonym names act as mediators that bridge the gap between different naming conventions (compounded vs. decompounded), allowing flexible query matching while maintaining the original database structure unchanged.
2Adaptability or versatility
If the system adds multiple synonym names for each geographic entity to accommodate different spelling forms, then query flexibility and user satisfaction improve, but data storage requirements and processing complexity increase
Solution Approach 1:
The synonym generation system applies universal rules that work across all geographic entity names regardless of their specific characteristics. The same algorithmic approach (identifying adjacent terms, concatenating them, validating against dictionary) is applied universally to generate both compounded and decompounded forms, making the system adaptable to any entity name without requiring entity-specific customization.
Solution Approach 2:
The system changes the parameter of name formatting by generating alternative representations (compounded and decompounded forms) of the same entity name. This parameter change allows the database to accommodate different user spelling preferences while storing the same underlying entity information, effectively increasing adaptability without proportionally increasing storage requirements.
3Measurement precision
If manual verification of synonym names is performed to ensure accuracy, then data quality and precision improve, but processing time and operational complexity increase
Solution Approach 1:
The system performs self-service validation by automatically checking generated synonym names against a dictionary to verify their legitimacy. This self-validation mechanism eliminates the need for manual verification while maintaining high accuracy, as the dictionary serves as an authoritative reference that automatically confirms or rejects potential synonym names.
Solution Approach 2:
The dictionary validation process provides immediate feedback on whether generated synonym names are legitimate words. This feedback mechanism allows the system to automatically filter out invalid synonyms and retain only accurate ones, ensuring high precision without manual intervention. The feedback loop continuously refines the synonym list based on dictionary verification results.
Data Source
AI summary
A synonym module analyzes a set of names for entities, such as names of streets and cities within a geographic database, and adds synonym names to the entities based on the analysis. More specifically, based on adjacencies of words within the names, the synonym module formulates a set of synonym rules specifying permissible ways of spelling a single word as multiple words (or vice-versa). The synonym module applies the synonym rules to the names, and if one of the rules matches a name, the synonym module adds an associated synonym name based on the name and on the rule matching the name. User queries for named entities in the database may then take into account not only the original names of the entities, but also any added synonym names.