Giovanni Da San Martino, Seunghak Yu, Alberto Barrón-Cedeño, Rostislav Petrov, Preslav Nakov. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint (EMNLP-IJCNLP). 2019.

We present the results and main findings of SemEval-2020 Task 11 on Detection Propaganda Techniques in News Articles. The task featured two subtasks. Subtask SI is about Span Identification: given a plain-text document, spot specific text fragments containing propaganda. TC Technique Classification: fragment, context full determine propaganda technique it uses, choosing from an inventory 14 possible techniques. attracted large number participants: 250 teams signed up to participate 44 made...

The reporting and the analysis of current events around globe has expanded from professional, editor-lead journalism all way to citizen journalism. Nowadays, politicians other key players enjoy direct access their audiences through social media, bypassing filters official cables or traditional media. However, multiple advantages free speech communication are dimmed by misuse media spread inaccurate misleading claims. These phenomena have led modern incarnation fact-checker --- a professional...

Although paraphrasing is the linguistic mechanism underlying many plagiarism cases, little attention has been paid to its analysis in framework of automatic detection. Therefore, state-of-the-art detectors find it difficult detect cases paraphrase plagiarism. In this article, we analyze relationship between and plagiarism, paying special which phenomena underlie acts them are detected by detection systems. With aim mind, created P4P corpus, a new resource that uses typology annotate subset...

Propaganda campaigns aim at influencing people's mindset with the purpose of advancing a specific agenda. They exploit anonymity Internet, micro-profiling ability social networks, and ease automatically creating managing coordinated networks accounts, to reach millions network users persuasive messages, specifically targeted topics each individual user is sensitive to, ultimately outcome on issue. In this survey, we review state art computational propaganda detection from perspective Natural...

Given the constantly growing proliferation of false claims online in recent years, there has been also a research interest automatically distinguishing rumors from factually true claims.Here, we propose general-purpose framework for fully-automatic fact checking using external sources, tapping potential entire Web as knowledge source to confirm or reject claim.Our uses deep neural network with LSTM text encoding combine semantic kernels task-specific embeddings that encode claim together...

Propaganda aims at influencing people's mindset with the purpose of advancing a specific agenda. Previous work has addressed propaganda detection document level, typically labelling all articles from propagandistic news outlet as propaganda. Such noisy gold labels inevitably affect quality any learning system trained on them. A further issue most existing systems is lack explainability. To overcome these limitations, we propose novel task: performing fine-grained analysis texts by detecting...

We present proppy, the first publicly available real-world, real-time propaganda detection system for online news, which aims at raising awareness, thus potentially limiting impact of and helping fight disinformation. The constantly monitors a number news sources, deduplicates clusters into events, organizes articles about an event on basis likelihood that they contain propagandistic content. is trained known sources using variety stylistic features. evaluation results standard dataset show...

End-to-end neural machine translation has overtaken statistical in terms of quality for some language pairs, specially those with large amounts parallel data. Besides this palpable improvement, networks provide several new properties. A single system can be trained to translate between many languages at almost no additional cost other than training time. Furthermore, internal representations learned by the network serve as a semantic representation words -or sentences- which, unlike standard...

In the context of investigative journalism, we address problem automatically identifying which claims in a given document are most worthy and should be prioritized for fact-checking.Despite its importance, this is relatively understudied problem.Thus, create new corpus political debates, containing statements that have been fact-checked by nine reputable sources, train machine learning models to predict fact-checking, i.e., model as ranking task.Unlike previous work, has looked primarily at...

We present the shared task on Fine-Grained Propaganda Detection, which was organized as part of NLP4IF workshop at EMNLP-IJCNLP 2019. There were two subtasks. FLC is a fragment-level that asks for identification propagandist text fragments in news article and also prediction specific propaganda technique used each such fragment (18-way classification task). SLC sentence-level binary asking to detect sentences contain propaganda. A total 12 teams submitted systems task, 25 did so 14...

Israa Jaradat, Pepa Gencheva, Alberto Barrón-Cedeño, Lluís Màrquez, Preslav Nakov. Proceedings of the 2018 Conference North American Chapter Association for Computational Linguistics: Demonstrations. 2018.

Community Question Answering (cQA) forums are very popular nowadays, as they represent effective means for communities around particular topics to share information. Unfortunately, this information is not always factual. Thus, here we explore a new dimension in the context of cQA, which has been ignored so far: checking veracity answers questions cQA forums. As problem, create specialized dataset it. We further propose novel multi-faceted model, captures from answer content (what said and...

Alberto Barrón-Cedeño, Daniele Bonadiman, Giovanni Da San Martino, Shafiq Joty, Alessandro Moschitti, Fahad Al Obaidli, Salvatore Romeo, Kateryna Tymoshenko, Antonio Uva. Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016). 2016.