Market-oriented job skill valuation with cooperative composition neural network.

Area of Science:

• Data Science
• Artificial Intelligence
• Human Resources Analytics

Background:

• Assessing job skill value is crucial for talent management but lacks quantitative methods.
• Existing approaches often fail to capture the market-oriented value of skills.

Purpose of the Study:

• To develop a data-driven solution for assessing job skill value from a market perspective.
• To propose a novel neural network model for skill value quantification and salary prediction.

Main Methods:

• Formulated job skill value assessment as a Salary-Skill Value Composition Problem.
• Developed an enhanced neural network, the Salary-Skill Composition Network (SSCN), utilizing massive job postings.
• Leveraged contextual job information and skill composition to determine skill value.

Main Results:

• The proposed SSCN model successfully assigns meaningful values to individual job skills.
• SSCN demonstrates superior performance compared to benchmark models in job salary prediction.
• The model effectively captures the market-oriented value of skills.

Conclusions:

• The data-driven approach using SSCN provides a quantitative method for job skill valuation.
• This methodology enhances talent selection and retention by offering objective skill assessments.
• The findings have significant implications for HR analytics and the labor market.