Constructing Domain-Specific Knowledge Graphs From Text: A Case Study on Subprime Mortgage Crisis

This research paper details a novel methodology for constructing a domain-specific Knowledge Graph (KG) from unstructured text data, exemplified by a case study on the subprime mortgage crisis. The authors present a five-phase approach – specification, conceptualization, formalization, integration, and augmentation – to transform unstructured financial news articles from the MEANTIME corpus into a structured KG. This framework enables the extraction of valuable insights, revealing trends, correlations, and complex relationships among companies, market movements, and economic indicators. The KG's efficacy is demonstrated through its ability to answer complex queries related to the subprime mortgage crisis, highlighting its potential as a powerful tool for knowledge representation and decision-making in the financial domain.