Entrepôts, Représentation et Ingénierie des Connaissances
Publications du laboratoire

Recherche approfondie

par Année
par Auteur
par Thème
par Type
- Link Prediction with Mutual Attention for Text-Attributed Networks doi link

Auteur(s): Brochier R., Guille A., Velcin J.

Conference: International Workshop on Deep Learning for Graphs and Structured Data Embedding (San Francisco, US, 2019-05-13)

Texte intégral en Openaccess : arxiv

Ref HAL: hal-02057120_v1
Ref Arxiv: 1902.11054
DOI: 10.1145/3308560.3316587
Ref. & Cit.: NASA ADS
Exporter : BibTex | endNote

In this extended abstract, we present an algorithm that learns a similarity measure between documents from the network topology of a structured corpus. We leverage the Scaled Dot-Product Attention, a recently proposed attention mechanism, to design a mutual attention mechanism between pairs of documents. To train its parameters, we use the network links as supervision. We provide preliminary experiment results with a citation dataset on two prediction tasks, demonstrating the capacity of our model to learn a meaningful textual similarity.