A knowledge graph to interpret proprietary data in the seismic industry

Efficiently conducting seismic hazard assessment and retrospective testing of seismic prediction strategies relies on the integration of proprietary seismic data into the forecasting analysis and decision process, however, the large volume and structural diversity of proprietary seismic data, and the fact that related knowledge is stored in multiple databases, throw a stumbling block to data integration. Hence, this paper introduce the Seismic Knowlee Graph (SKG), a flexible and powerful platform currently containing nearly 34,043 nodes and 32,248 relationships, representing the relevant observed data, public databases. The platform contains a variety of proprietary seismic data in different formats and from different data sources, providing data support to realize the data modeling and business process processing of strong and impending earthquake prediction model, so as to build a new model of regional earthquake forecasting study. In this work, we perform data organization of relational databases, and after graph database modeling and data import, we build knowledge graph using HugeGraph, a kind of NoSQL graph database, which organized easily extensible architecture that can be easily scale to new nodes and relationships when generates new data . We use WebGIS technology to build intuitive interfaces to visual graphical databases for user interaction, querying, and roaming the SKG. We illustrate the possibility of using a graph database to build a knowledge graph for seismic forecasting, which establishes a seismic forecast base database, a distributed database system for proprietary earthquake data. This graph database of relational queries has the possibility to reveal new relationships between heterogeneous seismic data and metadata, and it can be demonstrated that the graph structure can provide an efficient and reliable data support service for testing the validity of technical solutions for determining the urgency of strong earthquake generation.