README.md

---
dataset_info:
  features:
  - name: PREMISE
    dtype: string
  - name: ORIGINAL_LABEL
    dtype: string
  - name: PREDICTED_E
    dtype: int64
  - name: PREDICTED_D
    dtype: int64
  - name: PREDICTED_R
    dtype: int64
  - name: PREDICTED_I
    dtype: int64
  splits:
  - name: train
    num_bytes: 2765321
    num_examples: 12489
  download_size: 1198189
  dataset_size: 2765321
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

# Dataset Card for "LOGICPOLITICS"

Description
- A contribution to the task of [Logical Fallacy detection](https://arxiv.org/abs/2202.13758)
- Repository: [A benchmark for Malicious Content Detection in Political Settings](https://github.com/ajaykc7/LOGICPOLITICS)
- Point of Contact: [Hieu Nguyen](mailto:hieutn@usc.edu), [Ajay Kc](ajaykc@usc.edu), [Manorama Patil](manorama@usc.edu)