This paper focuses on reconciling knowledge graphs generated from two text documents about similar events described differently. The proposed approach employs and extends MERGILO, a tool for reconciling knowledge graphs extracted from text, using word similarity and graph alignment. Our approach effectively handles events using Frame Embeddings and Frame Based Similarities. It is evaluated over a coreference resolution task.
Frame Embeddings for Event-Based Knowledge Reconciliation / Alam A, Reforgiato Recupero D, Mongiovì M, Gangemi A, Ristoski P. - ELETTRONICO. - (2017), pp. 13-18. (Intervento presentato al convegno ISWC 2017 tenutosi a Vienna nel 2017).
Frame Embeddings for Event-Based Knowledge Reconciliation
GANGEMI, ALDO
2017
Abstract
This paper focuses on reconciling knowledge graphs generated from two text documents about similar events described differently. The proposed approach employs and extends MERGILO, a tool for reconciling knowledge graphs extracted from text, using word similarity and graph alignment. Our approach effectively handles events using Frame Embeddings and Frame Based Similarities. It is evaluated over a coreference resolution task.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.