EGU25-14239, updated on 15 Mar 2025
https://doi.org/10.5194/egusphere-egu25-14239
EGU General Assembly 2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Thursday, 01 May, 10:45–12:30 (CEST), Display time Thursday, 01 May, 08:30–12:30
 
Hall X5, X5.148
GeoGPT: Transforming Paleontology with AI-Powered Data Extraction and Analysis
Zongyuan Xiang
Zongyuan Xiang
  • Zhejiang Lab, Research center for data hub and security, China (zongyuan.xiang@zhejianglab.org)

GeoGPT is a non-profit domain-specific Large Language Model for geosciences, trained based on open-source data. It provides an effective solution to the challenges of managing large data volumes, complex formats, and low efficiency in the utilization of books and papers in the field of paleontology. Its powerful data extraction capabilities will significantly enhance the efficiency of extracting, analyzing, and building databases for data of various formats, sizes, and origins. This enables scientists to construct online fossil datasets and empowers paleontologists to develop innovative tools such as paleontological classification assistants. Not only does this accelerate scientific research progress, but it also makes the acquisition and application of paleontological data, such as invertebrate fossils, more convenient, ultimately driving comprehensive progress in the field of paleontology.

How to cite: Xiang, Z.: GeoGPT: Transforming Paleontology with AI-Powered Data Extraction and Analysis, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-14239, https://doi.org/10.5194/egusphere-egu25-14239, 2025.