Relation Clustering in Narrative Knowledge Graphs
This work provides a valuable preprocessing step for semi-supervised approaches to knowledge graph extraction from literary texts, which is useful for researchers and practitioners working with narrative data.
This paper addresses the challenge of extracting structured information from literary texts by clustering relations between entities in narrative knowledge graphs. It uses SBERT to embed relational sentences and then clusters them, followed by BART for summarization and label extraction, showing preliminary success in detecting similar relations.
When coping with literary texts such as novels or short stories, the extraction of structured information in the form of a knowledge graph might be hindered by the huge number of possible relations between the entities corresponding to the characters in the novel and the consequent hurdles in gathering supervised information about them. Such issue is addressed here as an unsupervised task empowered by transformers: relational sentences in the original text are embedded (with SBERT) and clustered in order to merge together semantically similar relations. All the sentences in the same cluster are finally summarized (with BART) and a descriptive label extracted from the summary. Preliminary tests show that such clustering might successfully detect similar relations, and provide a valuable preprocessing for semi-supervised approaches.