A Prediction Model of Human Resources Recruitment Demand Based on Convolutional Collaborative BP Neural Network

Area of Science:

• Artificial Intelligence
• Machine Learning
• Data Science

Background:

• Traditional enterprise talent assessment is susceptible to subjective factors, leading to errors and reduced validity.
• Online labor markets require advanced models for efficient matching of supply and demand.
• Existing models for employer hiring decisions, reputation analysis, and applicant recommendations need improvement.

Purpose of the Study:

• To develop and analyze a prediction model for recruitment demand using a hybrid convolutional neural network (CNN) and BP neural network approach.
• To enhance enterprise management talent assessment by reducing subjectivity and improving accuracy.
• To create more reliable models for employer hiring decisions, applicant reputation evaluation, and project recommendations in online labor markets.

Main Methods:

• A convolutional neural network (CNN) combined with a BP neural network algorithm for recruitment demand prediction.
• Application of BP neural network technology for enterprise management talent assessment, leveraging its adaptive learning and feedback adjustment capabilities.
• Development of an applicant reputation evaluation model using multiplicative long- and short-term recurrent neural networks (mLSTM).
• Construction of a hybrid project recommendation model based on a conditional variational self-encoder (CVAE).

Main Results:

• The proposed employer hiring decision model, reputation analysis model, and applicant project recommendation model demonstrated more reliable performance than existing models.
• Experimental validation on the Freelancer dataset confirmed the effectiveness of the developed models.
• The integrated approach successfully addresses the limitations of subjective factors in talent assessment, reducing errors and enhancing accuracy.

Conclusions:

• The developed hybrid neural network model offers a robust solution for predicting recruitment demand and improving talent assessment.
• The research provides technical support for online labor market platforms to offer personalized, intelligent, and accurate services.
• The findings contribute to more efficient matching of labor supply and demand, benefiting both employers and applicants.