Model-Driven Candidate Sorting via Audio Cues

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The process of evaluating job candidates is time-consuming and costly, and subjective, often leading to inconsistent results due to human evaluators' biases, making it difficult to accurately predict candidates' performance and achievement outcomes.

Innovation Solution

A model-driven candidate-sorting tool that analyzes digital interview data, including audio, video, and user interaction metrics, to predict an achievement index for candidates, allowing for objective and efficient filtering and ranking of candidates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual evaluation of candidates is performed by human interviewers, then subjective assessment of candidate qualifications is achieved, but the process becomes time-consuming and costly

Engineering Contradiction:
Improveassessment accuracyVSAvoidevaluation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical system of human interviewers manually evaluating candidates with an automated computing system that uses algorithms to analyze interview responses, resume data, and other candidate information. This substitution eliminates the time-consuming manual review process while maintaining or improving assessment accuracy through consistent, bias-free evaluation criteria.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service evaluation by automatically processing and comparing candidate data without requiring human intervention for each evaluation. The computing system independently performs data collection, analysis, scoring, and ranking, freeing human reviewers from routine evaluation tasks while preserving their ability to make final hiring decisions based on automated recommendations.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If human evaluators review candidate responses individually, then detailed assessment of each candidate is possible, but side-by-side comparisons become difficult and tedious

Engineering Contradiction:
Improveresponse evaluation accuracyVSAvoidcomparison ease
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent merges multiple candidate evaluations into a unified automated process that simultaneously analyzes and compares all candidates. The system combines individual response assessments with side-by-side comparisons by processing all candidate data through the same evaluation algorithms, automatically generating comparative rankings that make it easy to identify top candidates without manual switching between individual reviews.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The evaluation system performs multiple functions simultaneously: it conducts individual response analysis, performs side-by-side comparisons, generates scores, creates rankings, and provides recommendations all through a single automated process. This multi-functional approach eliminates the need for separate manual comparison steps while maintaining detailed evaluation accuracy.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Loss of information

If interviews are reviewed linearly from beginning to end, then complete candidate profiles are assessed, but comparing responses to specific questions across candidates requires tedious reordering and cross-comparing

Engineering Contradiction:
Improvecandidate information completenessVSAvoidcross-comparison time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent transforms the linear review process into a multi-dimensional analysis by organizing candidate data into structured formats that enable simultaneous access across different dimensions. The system creates a matrix view where candidates are compared across multiple questions and attributes simultaneously, allowing evaluators to access complete candidate profiles while easily comparing specific responses across all candidates without linear navigation.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system performs preliminary organization and structuring of candidate data before the evaluation process begins. Interview responses are pre-tagged, indexed, and organized by question and candidate, allowing the automated system to quickly retrieve and compare specific responses across candidates without requiring time-consuming reordering during the review process.

Inventive Principle:
Principle #10Preliminary action

4Productivity

If automated interview response gathering is implemented, then efficiency is improved, but evaluation of responses still requires significant human effort

Engineering Contradiction:
Improveresponse gathering efficiencyVSAvoidresponse evaluation time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent extends the automation continuum by making the evaluation process continuous with the data gathering process. Instead of stopping automation at response collection and manually resuming for evaluation, the system continuously processes data through automated analysis algorithms that generate evaluations, scores, and rankings without human intervention, maintaining productivity gains throughout the entire hiring workflow.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3080761B1Model-driven candidate sorting based on audio cues
Publication Date: 2021.04.21 HIREVUE
  • EP3080761B1 patent drawingFigure 1
  • EP3080761B1 patent drawingFigure 2~3
  • EP3080761B1 patent drawingFigure 4

AI summary

Methods and systems for model-driven candidate sorting for evaluating digital interviews are described. In one embodiment, a model-driven candidate-sorting tool selects a data set of digital interview data for sorting. The data set includes candidate for interviewing candidates (also referred to herein as interviewees). The model-driven candidate-sorting tool analyzes the candidate data for the respective interviewing candidate to identify digital interviewing cues and applies the digital interview cues to a prediction model to predict an achievement index for the respective interviewing candidate. This is performed without reviewer input at the model-driven candidate-sorting tool. The list of interview candidates is sorted according the predicted achievement indices and the sorted list is presented to the reviewer in a user interface.