An article by our scientists entitled “Combining Recommender Systems and Language Models in Early Detection of Signs of Anorexia” has been published in open access. The work describes the approach that obtained the best result according to the ERDE metric in the international competition for early risk prediction on the Internet – eRisk 2024. Authors of the work: Oskar Riewe-Perła and Prof. Agata Filipowska.
Combining recommender systems and language models in early detection of signs of anorexia
Continued achievements in the global information verification competition
In 2024, scientists from the Department of Information Systems participated for the second time in the CheckThat! competition, part of the international CLEF conference, focused on identifying check-worthy texts. Leveraging transformer-based language models and cross-linguistic transfer learning techniques, the team secured two second-place rankings for English and Arabic submissions.
Combining multiple attack methods for effective adversarial text generation
An article by our scientists entitled “OpenFact at CheckThat! 2024: Combining Multiple Attack Methods for Effective Adversarial Text Generation” has been published in open access. The paper describes the approach that won first place in an international competition in the area of information credibility.
Participation in an international competition in the area of analysis of multi-author writing style
Scientists from the Department of Information Systems took part in an international competition for the analysis of multi-author writing style – 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.
Panel discussion at the Wikimania 2024 conference (video)
During the international conference Wikimania 2024, a panel discussion entitled “Keynote: Opening the Academia” took place, in which Dr. Włodzimierz Lewoniewski participated. The panel concerned the topic of open science and its various aspects. In particular, the challenges faced by academic institutions that try to implement the principles of open science and the obstacles that hinder these processes were discussed. The role of Wikimedia projects, such as Wikipedia, in the … Read More
Analysis of cryptocurrency exchanges using open data
The work of our scientists on the analysis of digital currency exchanges has been published on the Springer website. As part of the research, models for predicting the success and bankruptcy of cryptocurrency exchanges were developed, based on financial data and information collected from open Internet sources such as Wikipedia and Wikidata. The aim of the work was to identify key aspects that can support the risk analysis process and … Read More
Wikimania 2024: Wikipedia and Wikidata as sources for analyzing Americanization (video)
During the Wikimania 2024 conference, the results of scientific research on measuring Americanization in various regions of the world based on open data from Wikipedia and Wikidata were presented. The conducted research reveals differences in interest in American topics depending on the level of development of countries, offering a new perspective on global cultural dominance. At the same time, the analysis emphasizes the role of Wikipedia and Wikidata as tools … Read More
The best result in an international competition for early risk prediction on the Internet
Employees of the Department of Information Systems at the Poznań University of Economics and Business, Prof. Agata Filipowska and Oskar Riewe-Perła, took part in the eRisk 2024 competition. The aim of the competition was to develop a system for early recognition of people with anorexia based on their activity on social media platforms.
First place in an international competition in the field of information credibility
A team of scientists from the Department of Information Systems at the Poznań University of Economics and Business took part in the “CheckThat! 2024” organized as part of the international conference CLEF (Conference and Labs of the Evaluation Forum). The goal was to verify the robustness of popular text classification approaches used for credibility assessment problems.
OpenFact is again the most effective in detecting fake news
The National Center for Research and Development has completed the evaluation of the results of the second phase of the INFOSTRATEG program. The evaluation included the results achieved by three competing teams whose task is to develop solutions for detecting fake news using artificial intelligence.