Paper Title: A comprehensive construction of the domain ontology for stratigraphy

Author: Huiqing Xu, Yingying Zhao, Hao Huang, Shaochun Dong, Yukun Sh, Chunju Huang, Huaichun Wu, Zhiqi Qian, Qiang Fang, Huaguo Wen, Zhongtang Su, Shuang Dai, Ronghua Wang, Chao Li, Chao Sun, Junxuan Fan

https://doi.org/10.1016/j.gsf.2022.101461

Available Online 28 August 2022

Introduction: Stratigraphic knowledge forms the bedrock of geoscience, and harnessing the power of Knowledge Graphs through ontology becomes paramount, especially in the context of modern big-data techniques. This study embarks on the ambitious journey of attempting to construct ontologies that encapsulate the complex and diverse world of stratigraphy. This endeavor is the result of a collaborative, crowd-driven initiative, orchestrated by domain experts hailing from various stratigraphic subdisciplines.

Key Terminology Enumeration: The first step involves a meticulous enumeration of key terms drawn from authoritative references, which are then seamlessly integrated into the Geoscience Professional Knowledge Graphs (GPKGs) within the framework of the Deep-time Digital Earth Project. Throughout this process, the team diligently addresses semantic heterogeneities, employing both professional judgment and automatic detection of Homonyms within the GPKGs platform.

Hierarchy Creation: Subsequently, these terms are further categorized as either classes or properties and meticulously structured within a hierarchical framework, employing a top-down approach. This meticulous endeavor results in the creation of seven distinct ontologies, each corresponding to a major branch of stratigraphy: Lithostratigraphy, Biostratigraphy, Chronostratigraphy, Chemostratigraphy, Magnetostratigraphy, Cyclostratigraphy, and Sequence Stratigraphy.

Biostratigraphy Unveiled: Within this trove of ontologies, Biostratigraphy emerges as a particular focus, as the creation of a comprehensive biostratigraphic ontology represents a novel endeavor within the field. This newly minted ontology encompasses essential root classes, including Fossil, Biostratigraphic unit, Biostratigraphic horizon, which collectively contribute to the ultimate endeavor of dating and correlating strata, encapsulated within the root class: Biostratigraphic correlation.

Potential Applications: In conclusion, this study's contributions may well represent one of the most expansive attempts at creating ontologies within the domain of stratigraphy. Additionally, a rudimentary model of a semantic search engine is postulated, sparking discussions about potential applications for improved querying of stratigraphic references. This hinges on the semantic connections between classes established within the constructed ontologies, hinting at a promising avenue for future exploration.

This undertaking represents a bold step toward unraveling the intricacies of stratigraphy, though it acknowledges the many challenges and questions that lie ahead. The potential applications of this work hold the promise of enhancing the way we explore and understand the Earth's geological history.