Communication Dans Un Congrès Année : 2024

Document Classification via Stable Graph Patterns and Conceptual AMR Graphs

Résumé

This paper proposes an approach and an associated system based on pattern structures, aimed at the classification of documents represented as graphs. The representation of documents relies on Abstract Meaning Representation (AMR) document graphs. Given a set of AMR document graphs, the system learns characteristic graph patterns, that can be reused by an aggregate rule classifier to predict the class of a document. The selection of the most stable graph patterns is based on the gSOFIA algorithm and the ∆-stability measure. In the experiments, two document datasets are considered for validating the approach. The first includes documents belonging to 10 different newsgroups and the second contains sports news articles belonging to 5 topical areas. The results in terms of the macro-averaged F1 scores, are quite satisfactory and show that the approach is well-founded and useful.

Fichier principal
Vignette du fichier
Document_Classification_via_Stable_Graph_Patterns_and_Conceptual_AMR_Graphs_Camera_Ready_Version.pdf (347.27 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04974939 , version 1 (03-03-2025)

Licence

Identifiants

Citer

Eric George Parakal, Egor Dudyrev, Sergei Kuznetsov, Amedeo Napoli. Document Classification via Stable Graph Patterns and Conceptual AMR Graphs. CONCEPTS 2024 - 1st International Joint Conference on Conceptual Knowledge Structures, Universidad de Cádiz, Sep 2024, Cádiz, Spain. pp.286-301, ⟨10.1007/978-3-031-67868-4_19⟩. ⟨hal-04974939⟩
111 Consultations
168 Téléchargements

Altmetric

Partager

  • More