Search Result Selection via Clustering and Quality Scores

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users often receive search results tailored to their inferred location, which may not be relevant to their intended search query, as search engines prioritize location-specific results over global results without considering the geographic clustering and quality of search results.

Innovation Solution

A search engine generates both location-specific and global sets of search results, assigns scores based on clustering, responsiveness, and quality, and selects the primary results set based on these scores to ensure relevance, allowing users to receive search results that may be geographically distant from their inferred location if they are more relevant.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If search engines prioritize location-specific search results based on inferred user location, then the search results are more relevant to the user's current location, but the search results may not be relevant to the user's intended search query when the user is looking for global or different location results

Engineering Contradiction:

Improvelocation accuracyVSAvoidsearch intent accuracy

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent segments search results into location-specific sets and global sets, allowing the system to evaluate and select between them based on different criteria including geographic clustering scores and quality scores, rather than automatically prioritizing location-specific results

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of location relevance by introducing cluster scores that measure geographic clustering and quality scores that estimate search result quality, enabling the system to select between location-specific and global results based on these dynamic parameters rather than fixed location-based rules

Inventive Principle:

Principle #35Parameter changes

2Productivity

If search engines automatically select location-specific search results, then the system operation is simple and fast, but the search result quality and relevance to user intent may be compromised

Engineering Contradiction:

Improvesearch result generation speedVSAvoidsearch result relevance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent performs preliminary actions by generating both location-specific and global search result sets in advance, and pre-calculating their respective cluster scores and quality scores, so that the final selection can be made quickly based on pre-evaluated metrics

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces feedback mechanisms through cluster scores and quality scores that evaluate the geographic clustering and quality of search result sets, allowing the system to adaptively select between location-specific and global results based on these feedback metrics

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If search engines provide global search results without location filtering, then the search results cover a broader geographic area, but the search results may not be geographically clustered and thus less relevant to users seeking local information

Engineering Contradiction:

Improvesearch scopeVSAvoidgeographic clustering

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent makes the search result selection dynamic by introducing cluster scores that measure geographic clustering and quality scores that evaluate result quality, allowing the system to adaptively determine whether to prioritize location-specific or global results based on these dynamic assessments rather than fixed rules

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS11461336B2Selecting between global and location-specific search results

Publication Date: 2022.10.04 GOOGLE LLC

AI summary

Methods, system, and apparatus, including computer programs encoded on computer storage media, for selecting between location-specific and global search results. One method includes receiving a query and a location identifier identifying a geographic location, obtaining a location-specific set of search results responsive to the query and a global set of search results responsive to the query, and determining a cluster score for the global set of search results. The method further includes selecting either the global set of search results or the location-specific set of search results as a primary set of search results according to one or more scores for the location-specific and global sets of search results, wherein the one or more scores include the cluster score, and identifying one or more identified search results in the primary set of search results for presentation in response to the query.