Query Processing System Using Feedback Logs for Latent Word Discovery

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing search engine indexing methods, such as separate word indexing and word partitioning, face inefficiencies and inaccuracies, particularly in languages without natural word separators like Chinese, failing to distinguish between similar terms and resulting in low recall and accuracy.

Innovation Solution

A query processing system that uses user feedback logs to determine latent word meanings by analyzing historical query results and selection frequencies, generating a second set of query keywords that include latent meanings, thereby improving search accuracy and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If separate word indexing is used, then the search can cover more terms, but the efficiency is low and accuracy is poor

Engineering Contradiction:
Improvesearch coverageVSAvoidsearch efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent segments the indexing process into two distinct parts: word partitioning indexing (for exact phrase matching) and separate word indexing (for individual term matching). This segmentation allows each indexing method to operate independently on its suitable data, improving overall efficiency while maintaining comprehensive search coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension to search indexing by creating a dual-indexing structure that operates at different levels: phrase-level indexing and word-level indexing. This dimensional approach allows the system to handle both exact phrase searches and flexible term searches simultaneously without compromising efficiency or accuracy.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If separate word indexing is used, then more terms can be searched, but the accuracy is poor

Engineering Contradiction:
Improvesearch coverageVSAvoidsearch accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent segments the indexing process into two distinct parts: word partitioning indexing (for exact phrase matching) and separate word indexing (for individual term matching). This segmentation allows each indexing method to operate independently on its suitable data, improving overall efficiency while maintaining comprehensive search coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different indexing qualities to different search scenarios: exact phrase matching uses word partitioning indexing for high precision, while individual term matching uses separate word indexing for broad coverage. This local quality approach ensures that each indexing method is applied where it is most effective.

Inventive Principle:
Principle #3Local quality

3Measurement precision

If word partitioning search technique is used, then the accuracy is high, but the recall rate is low

Engineering Contradiction:
Improvesearch accuracyVSAvoidrecall rate
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments the indexing process into two distinct parts: word partitioning indexing (for exact phrase matching) and separate word indexing (for individual term matching). This segmentation allows each indexing method to operate independently on its suitable data, improving overall efficiency while maintaining comprehensive search coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent merges two indexing techniques (word partitioning indexing and separate word indexing) into a unified search system. This combination allows the system to leverage the high accuracy of word partitioning while simultaneously capturing the broad recall capabilities of separate word indexing, thus resolving the contradiction between accuracy and recall rate.

Inventive Principle:
Principle #5Merging (Combining)

4Adaptability or versatility

If hybrid indexing method is used, then the disadvantages of earlier methods are made up, but the accuracy is affected due to inability to distinguish similar terms

Engineering Contradiction:
Improvesearch coverageVSAvoidsearch accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent segments the indexing process into two distinct parts: word partitioning indexing (for exact phrase matching) and separate word indexing (for individual term matching). This segmentation allows each indexing method to operate independently on its suitable data, improving overall efficiency while maintaining comprehensive search coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different indexing qualities to different search scenarios: exact phrase matching uses word partitioning indexing for high precision, while individual term matching uses separate word indexing for broad coverage. This local quality approach ensures that each indexing method is applied where it is most effective.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS9430568B2Method and system for querying information
Publication Date: 2016.08.30 ALIBABA GROUP HOLDING LTD
  • US9430568B2 patent drawing
  • US9430568B2 patent drawing
  • US9430568B2 patent drawing

AI summary

Performing a query includes receiving a user input, determining a first set of query keywords based at least in part on the user input, obtaining, based on at least some of the first set of query keywords, a user feedback log that includes historical query results, determining a second set of query keywords based on the user feedback log, and making a query based on at least some of the second set of query keywords.