Author(s)

Lankun Chen

Corresponding Author

Lankun Chen

Abstract

This thesis examines the use of large language models to enhance recruitment by automatically extracting and analyzing regional recruitment requirements from the job descriptions. This study addresses the key weaknesses in current assessment tools, such as ambiguity in job descriptions, excessive length, and bias. This research uses Natural Language Processing (NLP), specifically regular expressions, and the Bidirectional Encoder Representations from Transformers (BERT) algorithm to extract relevant elements from job descriptions. It employs a Topic Modeling approach to identify differences in regional recruitment preferences. The research demonstrates that automated NLP methods can be used to generate customized questionnaires, reducing ambiguity in job descriptions and unconscious bias while speeding up the hiring process. The research reveals significant differences in the recruitment requirements of different regions and platforms. It provides insights into developing more efficient and equitable strategies for recruitment. The thesis contributes to the field by providing a practical solution for improving the recruitment process using advanced AI and NLP technology while highlighting the need to be transparent, customized, and cost-efficient. Future research directions will include expanding data sources and addressing biases. They will also enhance model interpretability and assess the long-term effects of AI on recruiting.

Keywords

Natural Language Processing; AI Technology; Human Resource Management