Inductive Light Graph Convolution Network for Text Classification Based on Word-Label Graph - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Inductive Light Graph Convolution Network for Text Classification Based on Word-Label Graph

Jinze Shi
  • Fonction : Auteur
  • PersonId : 1275181
Xiaoming Wu
  • Fonction : Auteur
  • PersonId : 1275182
Xiangzhi Liu
  • Fonction : Auteur
  • PersonId : 1275183
Wenpeng Lu
  • Fonction : Auteur
  • PersonId : 1275184
Shu Li
  • Fonction : Auteur
  • PersonId : 1275185

Résumé

Nowadays, Graph Convolution Networks (GCNs) have flourished in the field of text classification, such as Text Graph Convolution Network (TextGCN). But good performance of those methods is based on building a graph whose nodes consist of an entire corpus, making their models transductive. Meanwhile rich label information has not been utilized in the graph structure. In this paper, we propose a new model named Inductive Light Graph Convolution Networks (ILGCN) with a new construction of graph. This approach uses labels and words to build the graph which removes the dependence between an individual text and entire corpus, and let ILGCN inductive. Besides, we simplify the model structure and only remain the neighborhood aggregation, which is the most important part of GCNs. Experiments on multiple benchmark show that our model outperforms existing state-of-the-art models on several text classification datasets.
Fichier sous embargo
Fichier sous embargo
0 6 29
Année Mois Jours
Avant la publication
mercredi 1 janvier 2025
Fichier sous embargo
mercredi 1 janvier 2025
Connectez-vous pour demander l'accès au fichier

Dates et versions

hal-04178743 , version 1 (08-08-2023)

Licence

Paternité

Identifiants

Citer

Jinze Shi, Xiaoming Wu, Xiangzhi Liu, Wenpeng Lu, Shu Li. Inductive Light Graph Convolution Network for Text Classification Based on Word-Label Graph. 12th International Conference on Intelligent Information Processing (IIP), May 2022, Qingdao, China. pp.42-55, ⟨10.1007/978-3-031-03948-5_4⟩. ⟨hal-04178743⟩
12 Consultations
1 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More