# The ParlaSent Multilingual Training Dataset for Sentiment Identification in Parliamentary Proceedings

arXiv: https://arxiv.org/abs/2309.09783 · HTML: https://mlsanity.com/paper/2309.09783/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 76 references that are in the map.
- First posted on arXiv 18 Sep 2023.

## Builds on (most cited first, 9 of 9)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Trankit: A Light-Weight Transformer-based Toolkit for Multilingual Natural Language Processing](https://mlsanity.com/paper/2101.03289/) (2021)
- [Multilingual Twitter Sentiment Classification: The Role of Human Annotators](https://mlsanity.com/paper/1602.07563/) (2016)
- [Sentiment and position-taking analysis of parliamentary debates: A systematic literature review](https://mlsanity.com/paper/1907.04126/) (2019)
- [CLASSLA-Stanza: The Next Step for Linguistic Processing of South Slavic Languages](https://mlsanity.com/paper/2308.04255/) (2023)
- [Bayesian Optimization of Text Representations](https://mlsanity.com/paper/1503.00693/) (2015)
- [False-Friend Detection and Entity Matching via Unsupervised Transliteration](https://mlsanity.com/paper/1611.06722/) (2016)
- [The ParlaSent-BCS dataset of sentiment-annotated parliamentary debates from Bosnia-Herzegovina, Croatia, and Serbia](https://mlsanity.com/paper/2206.00929/) (2022)

## Built on by (most cited first, 6 of 6)

- [Language Models on a Diet: Cost-Efficient Development of Encoders for Closely-Related Languages via Additional Pretraining](https://mlsanity.com/paper/2404.05428/) (2024)
- [Umm... With Transformers? Insights from Filled Pause Use across Four Slavic Parliaments](https://mlsanity.com/paper/2607.05964/) (2026)
- [Improving Customer Service with Automatic Topic Detection in User Emails](https://mlsanity.com/paper/2502.19115/) (2025)
- [Affective Polarization across European Parliaments](https://mlsanity.com/paper/2508.18916/) (2025)
- [SkMTEB: Slovak Massive Text Embedding Benchmark and Model Adaptation](https://mlsanity.com/paper/2606.13647/) (2026)
- [Opinionated, Hesitant and Stressed: Three Studies of How Politicians Speak in Four Slavic Parliaments](https://mlsanity.com/paper/2608.30828/) (2026)

## Abstract (arXiv)

The paper presents a new training dataset of sentences in 7 languages, manually annotated for sentiment, which are used in a series of experiments focused on training a robust sentiment identifier for parliamentary proceedings. The paper additionally introduces the first domain-specific multilingual transformer language model for political science applications, which was additionally pre-trained on 1.72 billion words from parliamentary proceedings of 27 European parliaments. We present experiments demonstrating how the additional pre-training on parliamentary data can significantly improve the model downstream performance, in our case, sentiment identification in parliamentary proceedings. We further show that our multilingual model performs very well on languages not seen during fine-tuning, and that additional fine-tuning data from other languages significantly improves the target parliament's results. The paper makes an important contribution to multiple disciplines inside the social sciences, and bridges them with computer science and computational linguistics. Lastly, the resulting fine-tuned language model sets up a more robust approach to sentiment analysis of political texts across languages, which allows scholars to study political sentiment from a comparative perspective using standardized tools and techniques.