/TagGNN-SIGIR

This repository contains the code of our SIGIR 2020 paper "Item Tagging for Information Retrieval: A Tripartite Graph Neural Network based Approach"

Primary LanguagePython

TagGNN

This is Pytorch implementation for our SIGIR 2020 paper:

Kelong Mao, Xi Xiao, Jieming Zhu, Biao Lu, Ruiming Tang, Xiuqiang He. Item Tagging for Information Retrieval: A Tripartite Graph Neural Network based Approach. Paper in arXiv.

Introduction

In this work, we proposed TagGNN, a heterogeneous graph neural network for more accurate item tagging under information retrieval scenario.

Code

To reproduce the results of the main table:

bash run.sh