Abstract

Geological profiles serve as invaluable repositories of data for knowledge extraction and the identification of geological entities, constituting a fundamental resource in scientific investigation. By encapsulating the stratigraphic arrangement and rock characteristics, these profiles offer a visually compelling and efficacious representation of stratigraphy along the vertical axis. Within geological profiles, intricate semantic associations frequently emerge, connecting various objects such as strata, rocks, and minerals across two or more profiles. These relationships often unveil inherent correlations or evolutionary attributes among geological entities. Understanding the general semantic relations between geological objects by manually scrutinizing vast quantities of profiles proves arduous, if not impracticable. In response, this study proposes an innovative framework capable of automated comprehension of extensive geological profiles and their associated contextual textual information. Built upon a natural language processing model and employing image vectorization techniques, this framework enables extraction of semantic information encompassing geological profiles, semantic relationships, and properties. In this article, we propose the comprehensive architecture of the framework, delineating its individual modules and subsequently applying it within a case study. Moreover, we conduct exploratory visualizations utilizing a knowledge graph to depict the extracted semantic information concerning diverse geological objects. The proposed framework presents a promising approach for supporting large-scale content analysis in the realm of stratigraphic research, facilitating advancements in our understanding of the evolution of stratigraphy.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call