Scientists from the Department of Information Systems took part in an international competition for the analysis of multi-author writing style \u2013 PAN 2024. PAN is a series of scientific events related to stylometric analysis and forensic linguistics, organized during the CLEF 2024 conference. The competition task was to detect places where the author changes in a text written by several authors.<\/p>\n
The main goal was to improve the model’s ability to detect authorship changes by increasing the model’s sensitivity to characteristic style elements. The key issue was to check whether it is possible to recognize differences in writing style. The method developed by the OpenFact team consisted in introducing stylometric tags directly into the text. As a result of the work, an article “Team OpenFact at PAN 2024: Fine-Tuning BERT Models with Stylometric Enhancements<\/a>” was created, describing the approach used. Authors of the publication: Ewelina Ksi\u0119\u017cniak<\/a>, Prof. Krzysztof W\u0119cel<\/a>, Marcin Sawi\u0144ski<\/a>.<\/p>\n It is worth mentioning that this year, a team of scientists from our Department took first place in international competitions in the area of \u200b\u200binformation credibility<\/a> and in the area of \u200b\u200bearly risk prediction<\/a> on the Internet. In addition, in 2023, the OpenFact project team took first place<\/a> in the “CLEF-2023 CheckThat! Lab” competition – the best method for detecting sentences in English that require verification due to the possibility of misleading.<\/p>\n Department of Information Systems is currently implementing the OpenFact research project, within which it is developing tools for automatic detection of fake news in Polish. In July 2024, the results of the OpenFact project were rated as the best in Poland<\/a> by National Center for Research and Development for the second year in a row. Victory in the prestigious CheckThat! competition confirms that the achievements of the PUEB team are important on a global scale and that the methods developed by the OpenFact team achieve equally high effectiveness in other languages.<\/p>\n