Building Dynamic Knowledge Graphs from Text-based Games
arXiv:1910.09532v312 citations
Originality Synthesis-oriented
AI Analysis
This work addresses the challenge of dynamic knowledge graph construction from textual data, but it is preliminary and incremental in nature.
The authors tackled the problem of updating knowledge graphs from text by proposing a novel Seq2Seq architecture to generate KG operations, and introduced a new dataset with over 300k data points from text-based games.
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.