Real-Time Survey Group Balancing via Vector Similarity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Creating a diverse survey group with specific demographic and socio-economic profiles is challenging and often requires significant manual oversight, leading to inefficiencies and high costs due to overages, where not enough participants meet the desired criteria, resulting in rejected candidates and increased expenses.

Innovation Solution

A method that uses vector representations of candidate profiles and survey group histograms to calculate similarity values in real-time, determining whether candidates fit the desired demographic and socio-economic distribution, with a threshold-based acceptance process to automatically balance the survey group composition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual oversight is used to ensure desired distribution, then survey group diversity is improved, but process complexity and time consumption increase significantly

Engineering Contradiction:

Improvesurvey group distribution accuracyVSAvoidmanual oversight complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual oversight with an automated computer-based system that uses algorithms to evaluate candidate profiles, calculate similarity scores against desired distributions, and automatically select respondents. This substitution eliminates the need for human reviewers while maintaining or improving distribution accuracy through consistent application of selection criteria.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by automatically managing the entire survey group creation process without human intervention. The computer-based system independently evaluates candidates, determines eligibility based on profile matching, and constructs the survey group autonomously, freeing researchers from manual monitoring tasks.

Inventive Principle:

Principle #25Self-service

2Manufacturing precision

If more candidates are evaluated to find suitable respondents, then survey group quality is improved, but overages and costs increase

Engineering Contradiction:

Improverespondent profile matching accuracyVSAvoidnumber of rejected candidates

Core Design Contradiction:

Manufacturing precisionVSLoss of substance

Solution Approach 1:

The system performs preliminary evaluation of candidate profiles before full survey selection by calculating similarity scores against the desired distribution. This preliminary filtering identifies the most suitable candidates in advance, allowing the system to select only those who best match the target profile, thereby reducing the number of rejected candidates while maintaining high matching accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from the histogram representing current survey group composition to dynamically adjust candidate selection. By continuously monitoring the distribution of selected respondents and comparing it against the desired distribution, the system can optimize its selection criteria in real-time, improving matching accuracy without unnecessarily evaluating excess candidates.

Inventive Principle:

Principle #23Feedback

3Productivity

If survey group creation is done quickly, then productivity is improved, but distribution accuracy may deteriorate

Engineering Contradiction:

Improvesurvey group creation speedVSAvoiddemographic distribution accuracy

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The system maintains continuous operation by automatically evaluating candidates as they become available and immediately determining their eligibility based on pre-calculated similarity scores. This continuous process eliminates delays associated with manual review batches while maintaining distribution accuracy through consistent application of selection criteria throughout the entire candidate pool.

Inventive Principle:

Principle #20Continuity of useful action

Solution Approach 2:

By pre-calculating similarity scores and maintaining ready-to-use evaluation criteria, the system can rapidly assess new candidates without time-consuming analysis. This preliminary preparation enables quick decision-making while preserving distribution accuracy, as the automated system consistently applies the same rigorous matching standards regardless of processing speed.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10373180B2Creating a survey sample group according to a desired participant distribution in real time

Publication Date: 2019.08.06 PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD

AI summary

A balanced survey is automatically created in real time, that is, while potential survey participants are evaluated. A survey histogram is automatically re-balanced with each new entry accepted as a respondent. An individual fills out a questionnaire providing demographic and socio-economic data. A vector representation of that person, referred to as an entry, is created. A similarity value is calculated indicating the similarity between the vector representation and a histogram vector representing the make-up of the survey. The similarity value is calculated by taking a dot product of the entry vector representation and the histogram vector. The system then determines whether the similarity value is greater than a similarity threshold value. If the similarity value exceeds the threshold value, the entry vector is integrated into the histogram vector, that is, the individual becomes part of the survey group.