A Survey of Methods and Architectures for Attention-Based Sparse Graph Convolutional Neural Network-Based Forecast Model for Career Planning in Human Resource Management

Abstract

The rapid advancement of artificial intelligence and big data analytics has significantly transformed human resource management (HRM), particularly in career planning and workforce analytics. Traditional HR systems rely on rule-based or statistical approaches that often fail to capture complex relationships among employees, skills, and job roles. With the increasing availability of workforce data, graph-based machine learning methods have emerged as effective tools for modeling such interconnected information. Graph Convolutional Neural Networks (GCNs) enable the representation of employees, skills, and job positions as nodes, capturing relationships through graph structures to predict career paths and recommend development strategies. However, conventional GCNs treat neighboring nodes equally, limiting their ability to reflect the varying importance of relationships. Attention-based mechanisms address this limitation by assigning dynamic weights to nodes, improving prediction accuracy and interpretability. Additionally, sparse graph convolutional networks effectively handle incomplete and heterogeneous HR data by learning meaningful representations from limited information.

Recent studies highlight the effectiveness of attention-based and heterogeneous graph neural networks in job recommendation and career forecasting. These models enhance recruitment decision-making and talent management. This survey reviews recent advancements, provides comparative analysis, and identifies research gaps for developing intelligent HR analytics systems.