Crowd-Sourced Instability Detection in IoT Computing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Crowdsourcing faces challenges in maintaining the integrity of data recommendations due to biased or irrelevant inputs from users, which can compromise the accuracy and reliability of crowd-sourced data, especially when biased data persists.

Innovation Solution

A cognitive system implements intelligent crowd-sourced instability detection by using a crowd-sourced deviation opinion (CSDO) model to identify and filter recommendations with bias scores exceeding a central tendency value threshold, thereby maintaining data integrity and preventing biased data from obscuring overall results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If crowd-sourced data is collected from multiple users, then the quantity and diversity of recommendations increase, but the reliability and integrity of the data deteriorate due to biased or irrelevant inputs

Engineering Contradiction:

Improvequantity of recommendationsVSAvoidintegrity of data

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent extracts and removes biased or irrelevant recommendations from the crowd-sourced data using automated filtering mechanisms. The system identifies recommendations that deviate from central tendency values and excludes them from the final aggregated results, thereby maintaining data integrity while preserving the benefits of crowd-sourced quantity.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system implements feedback loops where user recommendations are continuously evaluated against established criteria and central tendency metrics. Recommendations that fail to meet quality thresholds trigger corrective actions, including filtering or requesting revisions, ensuring that only reliable data contributes to the final outcomes.

Inventive Principle:

Principle #23Feedback

2Loss of information

If all crowd-sourced recommendations are included in the results, then the completeness of data is improved, but the accuracy of overall results deteriorates due to biased data obscuring valid recommendations

Engineering Contradiction:

Improvecompleteness of dataVSAvoidaccuracy of results

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The system extracts and removes biased or irrelevant recommendations from the crowd-sourced data using automated filtering mechanisms. The system identifies recommendations that deviate from central tendency values and excludes them from the final aggregated results, thereby maintaining data integrity while preserving the benefits of crowd-sourced quantity.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system changes the parameter of data inclusion by dynamically adjusting which recommendations are included based on their deviation from central tendency values. Recommendations exceeding predefined bias thresholds are excluded, while those within acceptable ranges are included, optimizing the balance between completeness and accuracy.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If automated filtering mechanisms are implemented to remove biased data, then the reliability of crowd-sourced data is improved, but the device complexity increases

Engineering Contradiction:

Improvereliability of dataVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system implements self-service filtering mechanisms that automatically evaluate and filter recommendations without requiring manual intervention. The automated processes use predefined algorithms to identify and exclude biased data, reducing the need for complex manual review systems while maintaining high reliability standards.

Inventive Principle:

Principle #25Self-service

4Difficulty of detecting and measuring

If bias scores are calculated for each recommendation, then the ability to identify biased data is improved, but the loss of time increases due to additional processing requirements

Engineering Contradiction:

Improvedetection of biased dataVSAvoidprocessing time

Core Design Contradiction:

Difficulty of detecting and measuringVSLoss of time

Solution Approach 1:

The system performs preliminary calculations of central tendency values and establishes bias thresholds before processing individual recommendations. This preliminary action enables rapid bias score calculation for each recommendation by comparing against pre-computed reference values, significantly reducing processing time while maintaining accurate detection capabilities.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11348020B2Intelligent crowd-sourced instability detection in a computing environment

Publication Date: 2022.05.31 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments for crowd-sourced instability detection in an Internet of Things (IoT) computing environment by a processor. A plurality of recommendations from a plurality of crowd-sourced users associated with a social graph may be collected. Those of the plurality of recommendations having a bias score exceeding a central tendency value threshold may be identified and transformed according to one or more corrective actions.