Similarity

📘

Ready to use this data? See the Similarity reference documentation to get started.

Description

This model is powered by probability-based machine learning that uses job postings data. As such, it doesn't have a unique dataset but instead aggregates the various job postings to perform its calculations. The model gets updated quarterly to reflect updates in those underlying datasets. Similarity scores are derived from the various permutations of skill and occupation pairs. For example, lotspecocc-skill will take a LOT specialized occupation ID as input and return a list of related skills. The models that don't list two taxonomies in their name, such as lotocc and onet, will return similar items from their respective taxonomy.

Use Cases

Questions answered by this dataset:

• Given a certain skill, what are some related occupations or skills?
• Given a certain occupation, what are some related occupations or skills?

Metrics

• Similarity: the similarity is a score between 0 and 1 where 0 represents no similarity and 1 represents an exact match

Filters

• Skills
• Occupation (O*NET, SOC, LOT)