Link Prediction with Mutual Attention for Text-Attributed Networks

Abstract : 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.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02057120
Contributor : Robin Brochier <>
Submitted on : Tuesday, March 5, 2019 - 10:15:56 AM
Last modification on : Wednesday, April 3, 2019 - 1:12:13 AM

Links full text

Identifiers

Collections

Citation

Robin Brochier, Adrien Guille, Julien Velcin. Link Prediction with Mutual Attention for Text-Attributed Networks. International Workshop on Deep Learning for Graphs and Structured Data Embedding, May 2019, San Francisco, United States. ⟨10.1145/3308560.3316587⟩. ⟨hal-02057120⟩

Share

Metrics

Record views

30