Query Substitution Context Classification for Search Precision

Solution Overview

Problem

Search engines face challenges in identifying reliable substitution contexts for query revision, as bad contexts introduce noise and generate unreliable substitute terms, affecting the accuracy and quality of search results.

Innovation Solution

A method that classifies substitution contexts into good and bad categories based on scoring criteria, discarding bad contexts and selecting the best ones to improve the quality of substitution rules, involving the evaluation of context hierarchies and frequency of occurrence to determine the relevance of substitute terms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If all substitution contexts are collected and evaluated, then the completeness of substitution rules is improved, but the system complexity and data processing load increase

Engineering Contradiction:

Improvenumber of substitution contextsVSAvoidsystem complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent extracts and removes bad substitution contexts from the data stream before they can be processed further. By identifying and discarding contexts that do not meet quality thresholds (such as contexts with insufficient disambiguation capability or those that introduce noise), the system reduces the volume of data requiring processing while maintaining the quality of remaining substitution rules.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs preliminary evaluation of substitution contexts before incorporating them into the final substitution rule set. By pre-filtering and scoring contexts based on their quality metrics (disambiguation capability, relevance, noise levels), the system prepares only the most promising contexts for further processing, reducing overall system complexity.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If bad substitution contexts are included in the evaluation, then the coverage of substitution rules is improved, but the accuracy and reliability of substitute terms deteriorate

Engineering Contradiction:

Improvecoverage of substitution rulesVSAvoidreliability of substitute terms

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent extracts and removes bad substitution contexts from the data stream before they can be processed further. By identifying and discarding contexts that do not meet quality thresholds (such as contexts with insufficient disambiguation capability or those that introduce noise), the system reduces the volume of data requiring processing while maintaining the quality of remaining substitution rules.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system converts the potential harm of bad contexts into a benefit by using them as training data for the evaluation model. By analyzing why certain contexts are marked as bad and adjusting the scoring criteria accordingly, the system improves its ability to distinguish good from bad contexts in the future, thereby enhancing the reliability of substitution rules.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

3Manufacturing precision

If comprehensive evaluation of substitution contexts is performed, then the quality of substitution rules is improved, but the processing time and computational resources increase

Engineering Contradiction:

Improvequality of substitution rulesVSAvoidprocessing time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary evaluation of substitution contexts before incorporating them into the final substitution rule set. By pre-filtering and scoring contexts based on their quality metrics (disambiguation capability, relevance, noise levels), the system prepares only the most promising contexts for further processing, reducing overall system complexity.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent extracts and removes bad substitution contexts from the data stream before they can be processed further. By identifying and discarding contexts that do not meet quality thresholds (such as contexts with insufficient disambiguation capability or those that introduce noise), the system reduces the volume of data requiring processing while maintaining the quality of remaining substitution rules.

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS9483581B2Evaluation of substitution contexts

Publication Date: 2016.11.01 GOOGLE LLC

AI summary

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for evaluating substitute terms. One of the methods includes receiving a query having an original term and determining one or more substitution contexts for the original term, wherein a substitution context includes one or more context terms and an indication of a position in the query of the original term and the one or more context terms. The substitution contexts are classified into a first category or a second category based on a respective score of each substitution context. The original term is associated with one or more substitution contexts in the first category.