Towards a Person-Job-Fit Recruitment: Job Prediction with Deep Neural Networks Based on Various Pre-trained Models

This research paper aims to explore job prediction by employing various deep neural network models, namely TextCNN, CNN-LSTM, TextRNN, and RCNN. The study utilizes diverse pre-trained word embeddings and a comprehensive dataset consisting of 350,000 distinct job descriptions sourced from multiple online job search platforms. The objective is to efficiently match individuals with suitable job roles based on their unique skill sets. These neural network models are implemented and fine-tuned using a unique dataset gathered from various online job portals in Germany. Moreover, we introduce an innovative ensemble model that integrates the strengths of the individual deep learning architectures to optimize performance on this task. The proposed ensemble model showcases superior performance over standalone models, achieving an F1-Score of 89% on an 89-class classification task. Additionally, we conduct a comprehensive analysis of the experimental results to deepen our understanding of this challenge and provide a robust foundation for developing more effective solutions in the future. Sophisticated models for person-job-fit, as explored in this study, is integral to refining talent acquisition and management strategies. The promising results obtained underscore the potential of advanced AI techniques in revolutionizing human resource management and tailoring career paths to individual skill sets.