SCITE

SCITE annotates causal relations in scientific texts at the token level, providing BIO-tagged cause and effect spans for training sequence labelling models in the scientific domain.

Overview

Domain	Science
Year	2021
Sentences	5,236

Tasks & Splits

Task	Train	Test
Causality Detection	4,450	786
Causal Event Candidate Extraction	4,450	786
Causality Identification	4,450	786

Data Explorer

Causality Detection

Causal Event Candidate Extraction

Causality Identification

Citation

@article{li:2021,
  title   = {Causality Extraction Based on Self-Attentive {{BiLSTM-CRF}} with Transferred Embeddings},
  author  = {Li, Zhaoning and Li, Qi and Zou, Xiaotian and Ren, Jiangtao},
  year    = 2021,
  journal = {Neurocomputing},
  volume  = {423},
  pages   = {207--219},
  doi     = {10.1016/J.NEUCOM.2020.08.078},
  urldate = {2024-07-22}
}