Process Extraction
Process extraction focuses on automatically transforming unstructured textual data into structured formats, primarily aiming to reduce the time and cost associated with manual data processing. Current research emphasizes leveraging machine learning, particularly deep learning models like U-Nets and transformers, along with large language models (LLMs), to achieve this extraction from diverse sources including scientific publications, clinical trials, and social media. This field is crucial for advancing various domains, from accelerating scientific discovery through automated literature analysis to improving business process management and enhancing the efficiency of clinical research.
Papers
General-Purpose vs. Domain-Adapted Large Language Models for Extraction of Structured Data from Chest Radiology Reports
Ali H. Dhanaliwala, Rikhiya Ghosh, Sanjeev Kumar Karn, Poikavila Ullaskrishnan, Oladimeji Farri, Dorin Comaniciu, Charles E. Kahn
Material Palette: Extraction of Materials from a Single Image
Ivan Lopes, Fabio Pizzati, Raoul de Charette