CycleGT: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training.
Qipeng GuoZhijing JinXipeng QiuWeinan ZhangDavid WipfZheng ZhangPublished in: CoRR (2020)
Keyphrases
- text mining
- directed graph
- text generation
- information retrieval
- semi supervised
- text retrieval
- graph mining
- connected components
- database
- graph theory
- free text
- syntactic categories
- graph theoretic
- graph representation
- textual data
- bipartite graph
- web documents
- training samples
- supervised learning
- graph matching
- random walk
- spanning tree
- feature space
- information extraction
- natural language generation
- graph data
- graphical models
- graph partitioning
- graph model
- unsupervised learning
- directed acyclic graph
- graph structure
- text documents