Back to results
US 12,755,249 B2Granted
Automated recruitment management method and system using an AI-based candidate transparent progress tracker
AssigneeObra Inc
InventorsRobb Lifferth, Paul Ahlstrom
Life of this patent
Estimating the term…
Abstract
A method and a system are disclosed for providing an Artificial Intelligence (AI) based recruitment management, the method comprising. An AI-based tracker module having a graphical user interface is configured for receiving a plurality of job applications from a plurality of candidates via respective user devices. The tracker is configured to predict the hiring probability of a candidate based on: a compensation viability score, a speed score, a skill sets of the candidate. The tracker is further configured to predict time to fill a position by the candidate, based on one or more parameters including location, salary, and skill set.
Claims14
A system for providing artificial intelligence (AI)-based recruitment management, the system comprising:
- a plurality of user devices, each of the plurality of user devices operating, via a server, an AI-based tracker module having a graphical user interface and a machine learning engine, configured to:
- receive a plurality of job applications from a plurality of candidates via respective user devices;
- calculate a compensation viability score (CVS) for each of the received job applications to determine a compensation range for a particular job position, wherein the CVS is calculated using a normalization algorithm based on candidate-provided salary expectations and stored market compensation data;
- analyse at least one job application of the received plurality of job applications, in an event the calculated CVS-is within the compensation range;
- for the at least one analyzed job application, apply a trained machine learning model to generate a speed score (SPD score) associated with candidate phase transitions and historical application processing times for similar positions, and dynamic candidate metrics;
- assign one or more skill set scores to the at least one job application being analysed;
- schedule one or more interviews of a candidate of the plurality of candidates to be conducted by at least one hiring manager;
- update the one or more skill set scores of the at least one job application based on each of the one or more interviews conducted, wherein the one or more skill set scores include:
- a skilled and experience score (SES) calculated by obtaining the highest salary expectation match among all active candidates and calculating a distance between the candidate and the highest salary expectation match;
- and
- a people, ethics & culture soft skills (PECS) score calculated by getting highest PECS match, among all active candidates and calculating a distance between the candidate and highest PECS match;
- fill the job position by hiring the candidate based on corresponding updated skill set scores and the calculated probability of hiring;
- and
- updating the trained machine learning model based on changes in the skill set scores, the speed score, and metrics derived from hiring for previously analyzed job applications, wherein updating the trained machine learning model comprises refining predictive accuracy of the machine learning model by comparing outcomes of previously analyzed job applications against baseline data generated from Similar Position Comparisons (SPCs) based on job variables including at least the SES, CVS, and geographical location, thereby improving a prediction of hiring for subsequently received job applications over a period of time without requiring manual recalibration of the machine learning.
11 more elements
A method for providing Artificial Intelligence (AI) based recruitment management, the method comprising:
- configuring a plurality of user devices, each of the plurality of user devices operating, via a server, an AI-based tracker module having a graphical user interface and a machine learning engine configured for:
- receiving a plurality of job applications from a plurality of candidates via respective user devices;
- calculating a compensation viability score (CVS) for each of the received job applications to determine a compensation range for a particular job position, wherein the CVS is calculated using a normalization algorithm based on candidate-provided salary expectations and stored market compensation data;
- analyzing at least one job application of the received plurality of job applications, in an event the calculated CVS score is within the compensation range;
- for the at least one analyzed job application, applying a trained machine learning model to generate a speed score (SPD score) associated with candidate phase transitions and historical application processing times for similar positions, and dynamic candidate metrics;
- assigning one or more skill set scores to the at least one job application being analyzed;
- scheduling one or more interviews of a candidate of the plurality of candidates to be conducted by at least one hiring manager;
- and
- updating the one or more skill set scores of the at least one job application based on each of the one or more interviews conducted, wherein the one or more skill set scores include:
- a skilled and experience score (SES) calculated by obtaining the highest salary expectation match among all active candidates and calculating a distance between the candidate and the highest salary expectation match;
- and
- a people, ethics & culture soft skills (PECS) score calculated by getting highest PECS match, among all active candidates and calculating a distance between the candidate and highest PECS match;
- filling the job position by hiring the candidate based on the updated skill set scores and the calculated probability of hiring;
- and
- updating the trained machine learning model based on changes in the skill set scores, the speed score, and metrics associated with hiring for previously analyzed job applications, wherein updating the trained machine learning model comprises refining predictive accuracy of the machine learning model by comparing outcomes of previously analyzed job applications against baseline data generated from Similar Position Comparisons (SPCs) based on job variables including at least the SES, CVS, and geographical location, thereby improving a prediction of hiring for subsequently received job applications over a period of time without requiring manual recalibration of the machine learning model.
12 more elements
Dependencies are inferred from the claim text.