Enhancing Talent Employment Insights Through Feature Extraction with LLM Finetuning
Karishma Thakrar, Nick Young
TL;DR
The paper tackles the challenge of extracting nuanced, machine-interpretable signals from unstructured job postings to support labor market analytics. Its approach combines semantic chunking, retrieval-augmented generation, and fine-tuned DistilBERT models, with Gemini 1.5 used to generate ground-truth labels on a 1.2 million postings dataset provided by AdeptID. Key contributions include a robust end-to-end pipeline for identifying remote availability, remuneration structure, and education/experience requirements, along with detailed per-variable evaluations and a discussion of limitations and scalability. The work offers a scalable framework for delivering richer insights to employers and job seekers and can be extended to additional features and deployment contexts.
Abstract
This paper explores the application of large language models (LLMs) to extract nuanced and complex job features from unstructured job postings. Using a dataset of 1.2 million job postings provided by AdeptID, we developed a robust pipeline to identify and classify variables such as remote work availability, remuneration structures, educational requirements, and work experience preferences. Our methodology combines semantic chunking, retrieval-augmented generation (RAG), and fine-tuning DistilBERT models to overcome the limitations of traditional parsing tools. By leveraging these techniques, we achieved significant improvements in identifying variables often mislabeled or overlooked, such as non-salary-based compensation and inferred remote work categories. We present a comprehensive evaluation of our fine-tuned models and analyze their strengths, limitations, and potential for scaling. This work highlights the promise of LLMs in labor market analytics, providing a foundation for more accurate and actionable insights into job data.