metadata
dataset_info:
features:
- name: text
dtype: string
- name: label_text
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 283193
num_examples: 1000
download_size: 158119
dataset_size: 283193
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- ko
tags:
- finance
pretty_name: FinSent-CLS-ko
size_categories:
- 1K<n<10K
FinSent-CLS-ko
- A task to classify Korean financial news sentences into Positive/Negative/Neutral categories.