matus's picture
Update README.md
44d2e39
|
raw
history blame
1.86 kB
metadata
language:
  - sk
tags:
  - twitter
license: cc
datasets:
  - dataset1
metrics:
  - f1
widget:
  - text: >-
      Najkrajšia vianočná reklama: Toto milé video vám vykúzli čarovnú
      atmosféru: Vianoce sa nezadržateľne blížia.
  - text: A opäť sa objavili nebezpečné výrobky. Pozrite sa, či ich nemáte doma

Sentiment Analysis model based on SlovakBERT

This is a sentiment analysis classifier based on SlovakBERT. The model can distinguish three level of sentiment:

  • -1 - Negative sentiment
  • 0 - Neutral sentiment
  • 1 - Positive setiment

The model was fine-tuned using Slovak part of Multilingual Twitter Sentiment Analysis Dataset [Mozetič et al 2016] containing 50k manually annotated Slovak tweets. As such, it is fine-tuned for tweets and it is not advised to use the model for general-purpose sentiment analysis.

Results

The model was evaluated in our paper [Pikuliak et al 2021, Section 4.4]. It achieves 0.670.67 F1-score on the original dataset and 0.580.58 F1-score on general reviews dataset.

Cite

@article{DBLP:journals/corr/abs-2109-15254,
  author    = {Mat{\'{u}}s Pikuliak and
               Stefan Grivalsky and
               Martin Konopka and
               Miroslav Blst{\'{a}}k and
               Martin Tamajka and
               Viktor Bachrat{\'{y}} and
               Mari{\'{a}}n Simko and
               Pavol Bal{\'{a}}zik and
               Michal Trnka and
               Filip Uhl{\'{a}}rik},
  title     = {SlovakBERT: Slovak Masked Language Model},
  journal   = {CoRR},
  volume    = {abs/2109.15254},
  year      = {2021},
  url       = {https://arxiv.org/abs/2109.15254},
  eprinttype = {arXiv},
  eprint    = {2109.15254},
}