Search Result Credibility Scoring via NLP and Emotion Analytics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users often encounter incorrect or biased search results online, making it difficult to distinguish fact from fiction, as conventional search services do not analyze the accuracy or objectivity of the content presented.

Innovation Solution

A computer system processes queries by analyzing search results using natural language processing and emotion analytics to score credibility, filtering out biased results and identifying consensus among search results to present only credible information to the user.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional search services return all relevant results, then users receive comprehensive information, but users cannot distinguish fact from fiction and are exposed to biased or incorrect content

Engineering Contradiction:

Improvecredibility of search resultsVSAvoidcomplexity of search processing system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary credibility analysis system that sits between the search engine and the user. This intermediary layer performs natural language processing and emotion analytics on search results to generate credibility scores, effectively mediating the information flow without requiring fundamental changes to the underlying search infrastructure.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The search result processing is segmented into distinct analytical components: natural language processing module, emotion analytics module, and credibility scoring module. Each component handles a specific aspect of credibility assessment, allowing the system to process information in manageable segments rather than as a monolithic complex task.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If analytics are performed on all search results to determine credibility, then accurate information is identified, but computational resources and network utilization increase

Engineering Contradiction:

Improveaccuracy of credibility assessmentVSAvoidcomputational resource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system applies partial action by performing full credibility analytics only on a subset of search results that meet certain criteria (e.g., top results, results from uncertain sources). For other results, simplified heuristics or pre-computed credibility metrics are used, reducing overall computational burden while maintaining adequate assessment accuracy for critical results.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent implements preliminary action through pre-computing credibility metrics for known sources and maintaining a database of source reputations. When new search results are generated, the system first checks against pre-computed data before performing full analytics, significantly reducing real-time computational requirements.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If multiple analytics including emotion analytics are applied to search results, then credible information is distinguished from biased content, but processing time increases

Engineering Contradiction:

Improveability to distinguish fact from biasVSAvoidsearch result processing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system implements periodic action by applying different levels of analytics at different stages. Initial filtering uses quick heuristics to eliminate obviously unreliable results, then emotion analytics and detailed NLP are applied periodically to the remaining subset, rather than to every result uniformly. This staged approach maintains reliability while reducing average processing time.

Inventive Principle:

Principle #19Periodic action

4Reliability

If search results are filtered to show only credible information, then users receive accurate results, but the quantity of information presented to users decreases

Engineering Contradiction:

Improveaccuracy of presented informationVSAvoidvolume of search results

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent applies local quality by presenting different quantities of information to different users based on their credibility scores and preferences. High-credibility results are presented in full detail, while lower-credibility results are either summarized, marked with credibility indicators, or excluded based on user settings. This allows the system to maintain accuracy while preserving information quantity through differential presentation.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS11347822B2Query processing to retrieve credible search results

Publication Date: 2022.05.31 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A computer system processes a query to retrieve credible search results. One or more data sources are searched to retrieve search results pertaining to the query. Analytics are performed on the search results to produce a score for the search results, wherein the score indicates credible information within the search results and the analytics include one or more from a group of natural language processing and emotion analytics. In response to the score indicating a lack of credible information within the search results, the search results are analyzed to determine a consensus within the search results, wherein the consensus indicates credible information within the search results. The credible information of the search results is presented. Embodiments of the present invention further include a method and program product for processing a query to retrieve credible search results in substantially the same manner described above.