From Text to Talent: A Pipeline for Extracting Insights from Candidate Profiles

TL;DR

The paper tackles efficient CV-to-multiple-vacancy matching by combining GPT-4 powered information extraction with OpenAI text embeddings and a heterogeneous graph framework. It introduces a four-stage pipeline that constructs a CV-based heterogeneous graph and trains GCN/RGCN models under ordinal and multi-label objectives, evaluated on a real-world anonymized dataset with $C=5461$ candidates and $S=39$ job openings. While GCN generally outperforms RGCN under the tested settings, the results demonstrate feasibility and potential for recruitment decision support in handling multi-vacancy scenarios with imbalanced data. The work highlights the practical impact of integrating language models with graph-based reasoning to streamline talent identification and placement in dynamic hiring environments, and proposes concrete paths for expanding to additional node types and tasks.

Abstract

The recruitment process is undergoing a significant transformation with the increasing use of machine learning and natural language processing techniques. While previous studies have focused on automating candidate selection, the role of multiple vacancies in this process remains understudied. This paper addresses this gap by proposing a novel pipeline that leverages Large Language Models and graph similarity measures to suggest ideal candidates for specific job openings. Our approach represents candidate profiles as multimodal embeddings, enabling the capture of nuanced relationships between job requirements and candidate attributes. The proposed approach has significant implications for the recruitment industry, enabling companies to streamline their hiring processes and identify top talent more efficiently. Our work contributes to the growing body of research on the application of machine learning in human resources, highlighting the potential of LLMs and graph-based methods in revolutionizing the recruitment landscape.