Automatic knowledge-graph creation from historical documents: The Chilean dictatorship as a case study
Loading...
Date
2024
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
We present our results regarding the construction of a knowledge graph from historical documents related to the Chilean dictatorship period (1973-1990). Our approach uses LLMs to automatically recognize entities and relations between them and resolve conflicts between these values. To prevent hallucination, the interaction with the LLM is grounded in a simple ontology with four types of entities and seven types of relations. To evaluate our architecture, we use a gold standard graph constructed using a small subset of the documents, and compare this to the graph obtained from our approach when processing the same set of documents. Results show that the automatic construction manages to recognize a good portion of all the entities in the gold standard and that those not recognized are explained mainly by the level of granularity in which the information is structured in the graph and not because the automatic approach misses an important entity in the graph. Looking forward, we expect this report to encourage work on other similar projects focused on enhancing research in humanities and social science. However, we remark that better evaluation metrics are needed to accurately fine-tune these types of architectures.