|
--- |
|
|
|
language: |
|
- hu |
|
tags: |
|
- text-classification |
|
license: gpl |
|
metrics: |
|
- accuracy |
|
widget: |
|
- text: "Jó reggelt! majd küldöm az élményhozókat:)." |
|
|
|
--- |
|
|
|
# Hungarian Sentence-level Sentiment Analysis model with huBERT |
|
|
|
- Pretrained model used: huBERT |
|
- Finetuned on Hungarian Twitter Sentiment (HTS) Corpus |
|
|
|
## Limitations |
|
|
|
- max_seq_length = 128 |
|
|
|
## Results |
|
|
|
| Model | HTS2 | HTS5 | |
|
| ------------- | ------------- | ------------- | |
|
| huBERT | **85.55** | 68.99 | |
|
| XLM-RoBERTa| 85.56 | 85.56 | |
|
|
|
## Citation |
|
If you use this model, please cite the following paper: |
|
|
|
``` |
|
@inproceedings {yang-bart, |
|
title = {Improving Performance of Sentence-level Sentiment Analysis with Data Augmentation Methods}, |
|
booktitle = {Proceedings of 12th IEEE International Conference on Cognitive Infocommunications (CogInfoCom 2021)}, |
|
year = {2021}, |
|
publisher = {IEEE}, |
|
address = {Online}, |
|
author = {{Laki, László and Yang, Zijian Győző}} |
|
pages = {417--422} |
|
} |
|
|
|
``` |
