Employee work environment analysis using natural language processing techniques and large-scale language models

Abstract

This study aimed to understand how employees feel about their work environment and the factors influencing those feelings, by analyzing reviews on job-switching websites. The data extracted using advanced natural language processing (BERT) revealed common factors affecting employee satisfaction, including workplace environment, career development, business growth, interpersonal relationships, and evaluation and rewards. Furthermore, by utilizing large language models to generate employee personas, the study identified specific challenges and motivations, suggesting the potential to infer employee characteristics and provide valuable insights for decision-making.