Torre’s Job-Matching Model | Years of experience

4.3. Years of experience

4.3.1.

Purpose

The skill experience ranker algorithm's purpose is to evaluate how many years of total experience the user has with all skills of the opportunity. In this case, we don’t only check if the job seeker can do the job like in the skill-proficiency factor, but we compare every user in how much relevant experience they have in each skill.

4.3.2.

Data

4.3.2.1.

Candidates

Candidates provide their experience data during various stages such as onboarding, profile editing, interactions with the recruiter bot, experience updates, and resume imports.

4.3.2.2.

Talent seekers

Talent seekers input their experience requirements for job roles during the job post creation or editing phase.

4.3.3.

Data validation

4.3.3.1.

Verifications

Candidate experiences and associated skills can be verified by other profiles, enhancing their credibility.

4.3.3.2.

Recommendations

Recommendations for skills can add further credibility to the claimed experiences.

4.3.3.3.

Through related experiences

The algorithm checks the duration of the candidate's experience with each skill.

4.3.3.4.

Skill-experience relevance

The relevance of each experience to the associated skill is evaluated.

4.3.4.

Algorithm validation

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

4.3.5.

Description

The algorithm assesses each required job skill against the candidate's experience in that skill. The process involves:

4.3.5.1.

Skill similarity

Skills are standardized and mapped to a normalized version.
String comparison techniques are used to assess similarity.
A skill graph or a data science/word2vec model is used to calculate similarity when there is no exact match.

4.3.5.2.

Experience comparison

The algorithm compares the candidate's years of experience in a skill against the job's requirement, with no upper limit on the score.

4.3.6.

Mathematical description

The mathematical formula for this process is as follows: Let:

J be the set of job skills.
G be the set of skills in the candidate's experience.
S(j) be a specific skill in the job skills set J.
S(g) be a specific skill in the candidate's experience set G.
Experience(S(j)) be the required years of experience for skill S(j) in the job.
Experience(S(j)) be the years of experience of skill S(g) for the candidate.

The total job matching score is calculated as:

Where:

Similarity_Score(S(j), S(g)) is defined as:
If exact string match → score = 1
Else
If the skills are connected in the skillgraph:
Multiply the weight of each edge until you reach the job skill node. The result of the multiplication is the weight.
Else:
Compute the vector for both skills using word2vec and compare both vectors to calculate the similarity instead.
Experience_score(S(j), S(g)) is defined as:

4.3.7.

Uncertainty

Uncertainty in this feature arises when a skill is missing from the candidate’s skill set. The system informs both the candidate and the talent seeker about this missing information, which is then reflected in the scoring.

4.3.8.

Alternatives

This experience-based approach differs from proficiency-based algorithms by focusing on the duration of skill usage rather than the depth of skill mastery. It offers an alternative perspective on a candidate's capability, emphasizing the time invested in skill development over the level of expertise achieved.