Approximate Search Query Scoring for Error-Tolerant Database Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Database search engines typically require exact input and fail to handle user queries with errors, leading to frustration and inefficiency, especially in consumer-oriented e-commerce websites where users may enter incorrect spellings or typing mistakes.
Innovation Solution
A method that identifies database features useful for searching, indexes them, and processes user queries by assigning rough scores based on phonetic and text analysis, followed by a rescoring process to provide relevant search results, even with query errors, using algorithms like Double Metaphone and edit distance calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the search engine requires exact input matches, then search precision is improved, but user frustration increases and search productivity decreases when users make typing errors or spelling mistakes
Solution Approach 1:
The system changes the parameter of string matching from exact equality to approximate matching based on edit distance and phonetic similarity. Instead of requiring query strings to exactly match database entries, the system calculates similarity scores using edit distance metrics and phonetic algorithms (如Double Metaphone) to determine matches, thereby maintaining precision while improving productivity
Solution Approach 2:
The patent introduces intermediary computational mechanisms (edit distance calculation, phonetic encoding) between the user query and database search. These intermediaries transform the exact matching problem into a similarity assessment problem, allowing the system to handle typos and spelling variations while still returning relevant results
2Measurement precision
If the search engine performs detailed analysis on all database records, then search accuracy is improved, but processing time increases significantly
Solution Approach 1:
The search process is segmented into multiple stages: first applying fast phonetic encoding to all records to generate rough scores, then applying more computationally intensive edit distance calculations only to records that pass the initial filtering. This segmentation allows detailed analysis to be applied selectively rather than universally, reducing overall processing time while maintaining accuracy
Solution Approach 2:
The system performs preliminary phonetic encoding and rough scoring on all database records before applying the more time-consuming edit distance calculations. This preliminary action filters out obviously non-matching records early, so that detailed analysis is only performed on a small subset of promising candidates, significantly reducing total processing time
Data Source
AI summary
A method for searching a database to produce search results from queries likely to contain errors. The process begins by identifying database features likely to be useful in searching, and those features are employed to index the database. After receiving a query from a user, the system develops a rough score for the query, by extracting features from the query, assigning match scores to query features matching database features; and assigning approximation scores to query features amenable to approximation analysis with database features. The rough score is used to identify identifying a set of database records for further analysis. Those records are then subjected to a more detailed rescoring process, based on correspondence between individual query elements and individual record elements, and between the query and the database record content, taken as a whole. Based on the rescoring process, output is provided to the user.