Search Result Filtering via Classifier Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Internet users are often inadvertently exposed to inappropriate or offensive content while searching online, such as pornography and violent material, which can be undesirable.

Innovation Solution

A method and system that utilize classifiers to analyze search queries and results, filtering out sensitive or offensive content by identifying protected classes of people and content types, thereby selecting and presenting search results that are non-offensive to users.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If search engines provide comprehensive search results, then information completeness is improved, but exposure to sensitive or offensive content increases

Engineering Contradiction:

Improveinformation completenessVSAvoidexposure to sensitive content

Core Design Contradiction:

Loss of informationVSObject-affected harmful factors

Solution Approach 1:

The search result set is segmented into multiple categories (safe results, sensitive results, offensive results) based on classifier analysis. Different segmentation rules are applied based on query classification, allowing comprehensive results to be delivered while protecting users from harmful content through structured categorization and selective filtering.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Classifiers act as intermediary components between the search engine and users. These classifiers analyze both queries and results, serving as a mediating layer that identifies and filters sensitive or offensive content before it reaches the user, thus maintaining information completeness while blocking harmful exposures.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If classifiers filter sensitive content from search results, then user protection is improved, but search result accuracy may deteriorate

Engineering Contradiction:

Improveuser protectionVSAvoidsearch result accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The filtering system dynamically adjusts its behavior based on query classification. Different filtering strategies are applied depending on whether the query is classified as safe, sensitive, or offensive. This dynamic approach ensures that user protection is maintained while minimizing unnecessary filtering of accurate search results in safe contexts.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes filtering parameters based on the classified query type. For safe queries, minimal filtering is applied to maintain accuracy. For sensitive or offensive queries, more aggressive filtering is applied to protect users. This parameter adjustment resolves the contradiction by adapting the level of protection to the specific search context.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10083237B2Protecting users from inappropriate sensitive or offensive search results

Publication Date: 2018.09.25 GOOGLE LLC

AI summary

A system and method for providing a search experience in which users are protected from exposure to inappropriate offensive or sensitive content is described. A search system may classify a search query and candidate search results obtained in response to the search query. Based on the classification of the search query and search results, the candidate search results may be modified to generate a set of search results presented to a user such that the presented search results do not include inappropriate sensitive or offensive content.