Authors: Zhiwei Ma, Javier E. Santo, Greg Lackey, Hari Viswanathan, Daniel O’Malley

Published on: May 08, 2024

Impact Score: 8.2

Arxiv code: Arxiv:2405.05438

Summary

• What is new: The use of large language models (LLMs) to extract information from historical documents of orphaned wells.
• Why this is important: Locating and plugging orphaned wells is essential for reducing environmental risks, but existing historical records are unstructured and outdated.
• What the research proposes: A computational approach leveraging LLMs, specifically Llama 2 models, to extract well location and depth from historical documents.
• Results: Achieved 100% accuracy in extracting information from clean, PDF-based reports, but accuracy drops to 70% with unstructured image-based records.

Technical Details

Technological frameworks used: Llama 2 models

Models used: Large language models (LLMs)

Data used: Dataset of 160 well documents

Potential Impact

Environmental risk management services, oil and gas industry, data digitization companies

Want to implement this idea in a business?

We have generated a startup concept here: EcoTrace.