Torre’s Job-Matching Model

4.4. Self-awareness

4.4.1.

Purpose

The purpose of the self-awareness factor is to validate if candidates understand what they have to offer in their profile and if the genome is providing value to the job posts they applied to.

4.4.2.

Data

For this factor, we take into account disqualifications and recommendations.

4.4.3.

Data validation

This factor by itself has the goal of validating the entire genome data so no further data validation is done.

4.4.4.

Algorithm validation

Daily and weekly metrics
Editorial reviews
Direct feedback from talent seekers
Direct feedback from candidates

4.4.5.

Description

The self-awareness score comprehends three different factors, all of them with their own algorithm.

$disqualificationsScore
$skillsRecommendationsScore

4.4.5.1.

DisqualificationScore

An initial positive event with score 1 and weight 1 is provided for every user. We take into account all the historical disqualifications / events for the user and their weights. These disqualification events can be negative, neutral and positive. Positive: Stronger candidates, getting hired. Neutral: Incomplete Genome, Screening Questions, Other. Negative: Skills required, Language required. We calculate a decaying factor based on how recent an event is We calculate the weighted average of all the disqualifications with their respective decaying factor, including the initial positive event. If the candidate has no disqualifications, we return null.

4.4.5.2.

SkillRecommendationScore

Having $skillsWithProficiencies as the candidate’s skills and proficiencies. Get $recommendedSkills with their proficiencies, from users that recommended the candidate by up to 3 skills per recommendation.

If the candidate has no recommended skills, return 0.5 as the score.

Filter $recommendedSkills to only use the ones that the candidate has.

If empty, return 0.5 as the score.

Calculate $awarenessCoefficient between the candidate’s $skillProficiency and $recommendedSkillProficiency Final score is 1.0 - the absolute value of the average of all the $awarenessCoefficient values.

If the user has no recommended skills, we return null.

4.4.5.3.

Total score

The final score is an average of the previously defined values, ranging between 0 and 1. If every score is null, we return 1.

4.4.6.

Mathematical description

It seems there was an issue with the formatting of the equations. Let's correct that for clarity:

4.4.6.1.

Disqualifications score

4.4.6.1.1.

Variables:

E: Set of historical disqualification events (including positive, neutral, negative) with their weights.
E_init: Initial positive event (score 1, weight 1).
decay(e): Decay function for event e based on recency.

4.4.6.1.2.

Formula:

4.4.6.2.

Skills recommendations score

4.4.6.2.1.

Variables:

S: Candidate’s skills with proficiencies.
R: Recommended skills with proficiencies from users.
awarenessCoefficient(s, r): Function to calculate the awareness coefficient for a candidate's skill s and recommended skill r.

4.4.6.2.2.

Formula:

4.4.6.3.

Total score

4.4.6.3.1.

Formula:

In this corrected representation, the equations are better formatted, and each case within the piecewise functions is clearly defined.