Torre’s Job-Matching Model | Location validation
3.6. Location validation
Location validation
3.6.1.
Purpose
The location-validation feature's purpose is to ensure that a user's location aligns with the geographical requirements of a job, based on self-reported locations and preferences on their profile.
3.6.2.
Data
3.6.2.1.
Candidates
The data for this factor is provided by the candidates themselves. Candidates can input and update their preferred work locations during the onboarding process of the platform, in the edit section while viewing their profiles, or through interactions with the recruiter bot. Additionally, we automatically determine a user’s likely location based on their IP address if they don’t specify a location preference.
3.6.2.2.
Talent seekers
Talent seekers can specify their location requirements in the job post during onboarding or while editing the job.
3.6.3.
Data validation
We validate the location data that candidates provide in several ways.
3.6.3.1.
Location verification
Candidates can verify their location on the platform through location services or manual address input. We confirm these locations for accuracy and feasibility relative to the job’s location requirements.
3.6.4.
Algorithm validation
  • Daily and weekly metrics
  • Editorial reviews
  • Direct feedback from talent seekers
  • Direct feedback from candidates
3.6.5.
Description
For every job, we compare the user’s preferred location with the job’s location requirements. If a user's preferred location does not align with the job's location requirements, we filter the user from the suggestions, and the user will score 0 for this factor.
3.6.6.
Mathematical description
Let:
  • L_r be the set of locations required for the job.
  • L_u be the set of preferred locations of the user.
Mathematical description
3.6.7.
Uncertainty
Uncertainty factors in this feature are primarily in “unspecified locations”. When a candidate's location preference is not specified, we do not assume it is misaligned but instead, we inform both the candidate and talent seeker about the missing information. This is then reflected in the score as an uncertainty range.
Fair artificial intelligence.
We’re committed to it.
Fair artificial intelligence.