Introducing JobTech's Proprietary Wage Prediction Model
This model predicts how much a job might pay on average and how wide the salary range could be. It evaluates job title, company name, required skills, and years of experience based on historical job advertisements with known salaries.
What Goes In
- Job Title & Company Name: The system interprets these contextually, similar to how humans read sentences.
- Skills: Converted into numerical format based on the model's learned skill database.
- Years of Experience (YOX): A numeric value indicating candidate experience level.
How It Works
The model employs two BERT models to separately interpret job title and company name, then integrates this understanding with skill and experience data. Years of experience are used to adjust the salary prediction — more experience usually means higher pay.
What It Predicts
- Average Wage: Expected earnings for the position.
- Wage Radius: The potential breadth of compensation variation.
The model generates predictions for both compensation level and salary range variability based on job characteristics and experience.