At the SEPLN conference in Jaén, Spain, the last result from the HOMADOS project was presented: a fine-grained propaganda detection solution. See publications for more information about the research and resources for the source code.
This concludes the project, but the publications and resources remain openly available for everyone. Big thanks to all that helped to make this effort possible!
IJCNN 2022 proceedings contain a new study from the HOMADOS group, showing how verification of a short claim can be facilitated by recommending relevant reliable sources. See Publications and Resources for more information.
Our article entitled “Using NLP to quantify the environmental cost and diversity benefits of in-person NLP conferences” has been accepted for the 60th Annual Meeting of the Association for Computational Linguistics (Findings of ACL 2022). See publications and resources for more information.
We participated in the shared task of “Detection of Persuasion Techniques in Texts and Images” at SemEval-2021 Task 6 and our solution was second best in subtask 2. See the publications on the task and our solution for more information.
Our article has been published in Information Processing & Management. It describes Credibilator, a browser extension performing automatic credibility assessment for the currently opened webpage. We also show how it can be used to alter users’ perception of text credibility.
Our research plan on fine-grained detection of propaganda techniques has been awarded 18,800 PLN within the Research Credits programme of Google Cloud Platform. Thanks, Google!
Our work has been recently covered by popular Polish media, such as Polish Press Agency, WP.PL or Dziennik Gazeta Prawna.
The HOMADOS crew has been joined by a new researcher, Konrad Kaczyński. See the current team here.
An article entitled “Multi-Word Lexical Simplification”, describing joint effort with Manchester Metropolitan University, will appear at the International Conference on Computational Linguistics in Barcelona, Spain (online). See publications and resources for more information.
Our work on credibility assessment will be presented by Piotr Przybyła at the seminar of the Linguistic Engineering Group (ZIL) at the Institute of Computer Science, Polish Academy of Sciences on the 9th of March.