The ParlaSent Multilingual Training Dataset for Sentiment Identification in Parliamentary Proceedings

TL;DR

The paper tackles the scarcity of domain-specific, multilingual sentiment resources for parliamentary discourse by introducing ParlaSent, a sentence-level sentiment dataset across seven languages, annotated with a six-level schema. It also presents XLM-R-parla, a domain-adapted multilingual transformer pre-trained on 1.72B words from parliamentary proceedings, and evaluates it against vanilla XLM-R models. Key findings show that parliamentary pre-training improves sentiment detection, the model generalizes well to unseen languages, and multilingual training generally benefits performance for target parliaments. This work provides a practical, scalable framework for cross-lingual political sentiment analysis and offers resources to social scientists for standardized comparative studies.

Abstract

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.