Knowledge-Driven Cross-Document Relation Extraction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jain, Monika, Mutharaju, Raghava, Singh, Kuldeep, Kavuluru, Ramakanth
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910491615428608
author Jain, Monika
Mutharaju, Raghava
Singh, Kuldeep
Kavuluru, Ramakanth
author_facet Jain, Monika
Mutharaju, Raghava
Singh, Kuldeep
Kavuluru, Ramakanth
contents Relation extraction (RE) is a well-known NLP application often treated as a sentence- or document-level task. However, a handful of recent efforts explore it across documents or in the cross-document setting (CrossDocRE). This is distinct from the single document case because different documents often focus on disparate themes, while text within a document tends to have a single goal. Linking findings from disparate documents to identify new relationships is at the core of the popular literature-based knowledge discovery paradigm in biomedicine and other domains. Current CrossDocRE efforts do not consider domain knowledge, which are often assumed to be known to the reader when documents are authored. Here, we propose a novel approach, KXDocRE, that embed domain knowledge of entities with input text for cross-document RE. Our proposed framework has three main benefits over baselines: 1) it incorporates domain knowledge of entities along with documents' text; 2) it offers interpretability by producing explanatory text for predicted relations between entities 3) it improves performance over the prior methods.
format Preprint
id arxiv_https___arxiv_org_abs_2405_13546
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Knowledge-Driven Cross-Document Relation Extraction
Jain, Monika
Mutharaju, Raghava
Singh, Kuldeep
Kavuluru, Ramakanth
Computation and Language
Information Retrieval
Relation extraction (RE) is a well-known NLP application often treated as a sentence- or document-level task. However, a handful of recent efforts explore it across documents or in the cross-document setting (CrossDocRE). This is distinct from the single document case because different documents often focus on disparate themes, while text within a document tends to have a single goal. Linking findings from disparate documents to identify new relationships is at the core of the popular literature-based knowledge discovery paradigm in biomedicine and other domains. Current CrossDocRE efforts do not consider domain knowledge, which are often assumed to be known to the reader when documents are authored. Here, we propose a novel approach, KXDocRE, that embed domain knowledge of entities with input text for cross-document RE. Our proposed framework has three main benefits over baselines: 1) it incorporates domain knowledge of entities along with documents' text; 2) it offers interpretability by producing explanatory text for predicted relations between entities 3) it improves performance over the prior methods.
title Knowledge-Driven Cross-Document Relation Extraction
topic Computation and Language
Information Retrieval
url https://arxiv.org/abs/2405.13546