%0 Conference Paper
%F Oral 
%T Link Prediction with Mutual Attention for Text-Attributed Networks
%+ Digital Scientific Research Technology (DSRT)
%+ Entrepôts, Représentation et Ingénierie des Connaissances (ERIC)
%A Brochier, Robin
%A Guille, Adrien
%A Velcin, Julien
%< avec comité de lecture
%B International Workshop on Deep Learning for Graphs and Structured Data Embedding
%C San Francisco, United States
%8 2019-05-13
%D 2019
%Z 1902.11054
%R 10.1145/3308560.3316587
%K attributed network
%K natural language processing
%K representation learning
%K link prediction
%Z Computer Science [cs]/Social and Information Networks [cs.SI]Conference papers
%X 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.
%G English
%L hal-02057120
%U https://hal.science/hal-02057120
%~ UNIV-LYON1
%~ UNIV-LYON2
%~ ERIC
%~ LYON2
%~ UDL
%~ UNIV-LYON