Warsaw, Poland, 2024-02-29

Parliamentary data analysis

Events NeuroLex

On the last day of February this year, a workshop was held to conclude a competition for the analysis of parliamentary data using modern natural language processing (NLP) techniques, including LLM models. The event was organized by the Institute of Computer Science Polish Academy of Sciences within the CLARIN-PL consortium. Courtesy of the Chancellery of the Sejm, the workshop was held on the grounds of the Polish Parliament, in the GI building.

The goal of the competition, organized by the IPI PAN, was to develop an original analysis (monolingual or multilingual, with obligatory participation of Polish data) presenting an interesting research problem, in which the main source of data was the multilingual corpus ParlaMint – a set of speeches from the parliaments of 26 countries and autonomous regions, recorded in a common format, annotated with linguistic metadata.

The competition was open to students and young scientists. The competition’s jury, which included scientists representing the University of Warsaw, Jagiellonian University, Collegium Civitas, the Institute of Political Studies of the Polish Academy of Sciences, SWPS University, the Institute of Computer Science of the Polish Academy of Sciences, the University of Wroclaw and the University of Lodz, awarded the following works:

  • Application for sentiment analysis of the statements of parliamentarians of the Republic of Poland – 1st place (Bartłomiej Klimowski),
  • Analysis of the dynamics of emotions in parliamentary debates about the war in Ukraine – 2nd place (Ewelina Gajewska, Konrad Kilian),
  • Gender-sensitive language in the Polish parliament: a corpus-based study of parliamentary debates in the ParaMint corpus – 3rd place (Aleksandra Tomaszewska, Anna Jamka).

More information about the competition can be found at: https://clarin.biz/news/konkurs-vol.5-parlamint