Enriching Documents with Compact, Representative, Relevant Knowledge Graphs

Abstract

A prominent application of knowledge graph (KG) is document enrichment. Existing methods identify mentions of entities in a background KG and enrich documents with entity types and direct relations. We compute an entity relation subgraph (ERG) that can more expressively represent indirect relations among a set of mentioned entities. To find compact, representative, and relevant ERGs for effective enrichment, we propose an efficient best-first search algorithm to solve a new combinatorial optimization problem that achieves a trade-off between representativeness and compactness, and then we exploit ontological knowledge to rank ERGs by entity-based document-KG and intra-KG relevance. Extensive experiments and user studies show the promising performance of our approach.

Cite

Text

Li et al. "Enriching Documents with Compact, Representative, Relevant Knowledge Graphs." International Joint Conference on Artificial Intelligence, 2020. doi:10.24963/IJCAI.2020/242

Markdown

[Li et al. "Enriching Documents with Compact, Representative, Relevant Knowledge Graphs." International Joint Conference on Artificial Intelligence, 2020.](https://mlanthology.org/ijcai/2020/li2020ijcai-enriching/) doi:10.24963/IJCAI.2020/242

BibTeX

@inproceedings{li2020ijcai-enriching,
  title     = {{Enriching Documents with Compact, Representative, Relevant Knowledge Graphs}},
  author    = {Li, Shuxin and Huang, Zixian and Cheng, Gong and Kharlamov, Evgeny and Gunaratna, Kalpa},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {1748-1754},
  doi       = {10.24963/IJCAI.2020/242},
  url       = {https://mlanthology.org/ijcai/2020/li2020ijcai-enriching/}
}