README.md

metadata

dataset_info:
  features:
    - name: tokens
      sequence: string
  splits:
    - name: train
      num_bytes: 803588
      num_examples: 12290
    - name: validation
      num_bytes: 12403
      num_examples: 221
    - name: test
      num_bytes: 16702
      num_examples: 257
  download_size: 229583
  dataset_size: 832693
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*
license: cc-by-4.0
language:
  - yue

This data is the subset of the Hong Kong Cantonese Corpus (HKCanCor) that has been re-segmented by the multi-tiered word segmentation scheme described in the following paper:

Charles Lam, Chaak-ming Lau, and Jackson L. Lee. 2024. Multi-Tiered Cantonese Word Segmentation. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 11993–12002, Torino, Italy. ELRA and ICCL.

This HuggingFace dataset is produced by splitting all separators (spaces, dashes, and pipes) in the original dataset to arrive at the most fine-grained segmentations.