publications

Association Rules Collaborative Annotation Data Mining on Graphs Dataset Retrieval Explainable Artificial Intelligence Graph Data Mining Graph Neural Networks In-Context Learning Information Extraction Knowledge Base Population Large Language Models Lexicon Induction Link Prediction Machine Learning Machine Learning on Graphs Natural Language Generation Natural Language Processing Ontology Engineering Pattern Mining Question Answering Relation Extraction Sequence Labeling Web Table Understanding 
Alle anzeigen
2024
 Blum, Moritz;  Nolano, Gennaro;  Ell, Basil;  Cimiano, Philipp
Investigating the Impact of Different Graph Representations for Relation Extraction with Graph Neural Networks Proceedings Article 
In: Proceedings of the Deep Learning and Linguistic Linked Data Workshop at LREC-COLING, Turin, 2024.
Abstract | Links | BibTeX | Schlagwörter: Graph Neural Networks, Machine Learning on Graphs
@inproceedings{2988690,

title = {Investigating the Impact of Different Graph Representations for Relation Extraction with Graph Neural Networks},

author = {Moritz Blum and Gennaro Nolano and Basil Ell and Philipp Cimiano},

url = {https://aclanthology.org/2024.dlnld-1.1.pdf},

year  = {2024},

date = {2024-01-01},

urldate = {2024-01-01},

booktitle = {Proceedings of the Deep Learning and Linguistic Linked Data Workshop at LREC-COLING},

address = {Turin},

abstract = {Graph Neural Networks (GNNs) have been applied successfully to various NLP tasks, particularly Relation Extraction (RE). Even though most of these approaches rely on the syntactic dependency tree of a sentence to derive a graph representation, the impact of this choice compared to other possible graph representations has not been evaluated. We examine the effect of representing text though a graph of different graph representations for GNNs that are applied to RE, considering, e. g., a fully connected graph of tokens, of semantic role structures, and combinations thereof. We further examine the impact of background knowledge injection from Knowledge Graphs (KGs) into the graph representation to achieve enhanced graph representations. Our results show that combining multiple graph representations can improve the model’s predictions. Moreover, the integration of background knowledge positively impacts scores, as enhancing the text graphs with Wikidata features or WordNet features can lead to an improvement of close to 0.1 in F1.},

keywords = {Graph Neural Networks, Machine Learning on Graphs},

pubstate = {published},

tppubtype = {inproceedings}

}

Schließen
Graph Neural Networks (GNNs) have been applied successfully to various NLP tasks, particularly Relation Extraction (RE). Even though most of these approaches rely on the syntactic dependency tree of a sentence to derive a graph representation, the impact of this choice compared to other possible graph representations has not been evaluated. We examine the effect of representing text though a graph of different graph representations for GNNs that are applied to RE, considering, e. g., a fully connected graph of tokens, of semantic role structures, and combinations thereof. We further examine the impact of background knowledge injection from Knowledge Graphs (KGs) into the graph representation to achieve enhanced graph representations. Our results show that combining multiple graph representations can improve the model’s predictions. Moreover, the integration of background knowledge positively impacts scores, as enhancing the text graphs with Wikidata features or WordNet features can lead to an improvement of close to 0.1 in F1.
Schließen
https://aclanthology.org/2024.dlnld-1.1.pdf
Schließen