metadata

license: cc-by-4.0
task_categories:
  - text-classification
language:
  - fi
size_categories:
  - 10K<n<100K

Dataset Card for HunEmPoli_finnish

This dataset is a machine translated version of the HunEmPoli dataset set available here: https://osf.io/67zsf/?view_only=a23e5b6ba5ef443892a885a3f1d1d1e7
Details about the dataset can be found in the original paper.
The dataset was translated into Finnish using DeepL: https://www.deepl.com/translator

Uses

The dataset can be used to train an emotion analysis model.

Dataset Structure

The data fiels are:

text: A sentence from the Hungarian parliament.
label: Aclassification label, where 0 = neutral, 1 = fear, 2 = sadness, 3 = anger, 4 = disgust, 5 = success, 6 = joy, 7 = trust.
id: Anique identifier for each sentence. Numbering matches the row numbers in the original dataset.

Recommendations

The dataset is machine translated and, thus, might include mistranslations. The quality of the translation has not been verified. Make sure the data is suitable for your use case!

Citation

Please, cite the original work when using the data.

@ARTICLE{10149341, author={Üveges, István and Ring, Orsolya}, journal={IEEE Access}, title={HunEmBERT: A Fine-Tuned BERT-Model for Classifying Sentiment and Emotion in Political Communication}, year={2023}, volume={11}, number={}, pages={60267-60278}, keywords={Analytical models;Task analysis;Sentiment analysis;Dictionaries;Social sciences;Bit error rate;Data models;Emotion recognition;Fine-tuned BERT-model;huBERT;emotion analysis;sentiment analysis;political communication}, doi={10.1109/ACCESS.2023.3285536} }