Building Dynamic Knowledge Graphs from Text-based Games
Published in
arXiv.org(2019)External Links:
TL
TL;DR
This preliminary work proposes a novel Sequence-to-Sequence (Seq2Seq) architecture to generate elementary KG operations and introduces a new dataset for KG extraction built upon text-based game transitions.
Abstract
We are interested in learning how to update Knowledge Graphs (KG) from text. In this preliminary work, we propose a novel Sequence-to-Sequence (Seq2Seq) architecture to generate elementary KG operations. Furthermore, we introduce a new dataset for KG extraction built upon text-based game transitions (over 300k data points). We conduct experiments and discuss the results.
References14 items
1
Does Order Matter? An Empirical Study on Generating Multiple Keyphrases as a Sequence
2
CLUTRR: A Diagnostic Benchmark for Inductive Reasoning from Text
3
Playing Text-Adventure Games with Graph-Based Deep Reinforcement Learning
4
Generating Diverse Numbers of Diverse Keyphrases
5