Embedded Files
AWAKEN: content-Aware and netWork-Aware faKE News mitigation
AWAKEN: content-Aware and netWork-Aware faKE News mitigation
Principal Investigator: Ciprian-Octavian TRUICĂ
Principal Investigator: Ciprian-Octavian TRUICĂ
Project Partners: Adrian Paschke (Freie Universität Berlin / Fraunhofer FOKUS)
Project Partners: Adrian Paschke (Freie Universität Berlin / Fraunhofer FOKUS)
Funding Agency: German Academic Exchange Service (DAAD)
Funding Agency: German Academic Exchange Service (DAAD)
Grant no.: 57552334
Grant no.: 57552334
Overview
Overview
This research project addresses the profound shift in the media landscape, where digital networks have replaced traditional mass media, turning users into active participants and giving rise to the rapid, real-time spread of misinformation. Because this digital environment accelerates ideological polarization and the viral spread of fake news, the project proposes a holistic, content-aware, and network-aware system designed to track, detect, and actively counter deceptive content on social networks.
Key Objectives
Key Objectives
Advanced Detection: Develop new, highly accurate models tasked specifically with spotting and identifying fake news content.
Mitigation Strategy: Implement fresh strategies to stop the viral spread of misinformation once detected.
Innovation and Impact
Innovation and Impact
Most traditional tools look only at the text of an article or the network of people sharing it. AWAKEN innovates by creating a hybrid, content- and network-aware system that analyzes both the message and the medium simultaneously. By moving beyond simple detection into active mitigation, this research offers a real-world toolset to protect readers from ideological manipulation, reduce social polarization, and restore structural trust in digital journalism.
Publications
Publications
Journals
Journals
2022
2022
Erkut Erdem, Menekse Kuyu, Semih Yagcioglu, Anette Frank, Letiția Pârcălăbescu, Barbara Plank, Andrii Babii, Oleksii Turuta, Aykut Erdem, Iacer Calixto, Elena Lloret, Elena-Simona Apostol, Ciprian-Octavian Truică, Branislava Šandrih, Albert Gatt, Sanda Martinčić-Ipšic, Gábor Berend, Gražina Korvel. Neural Natural Language Generation: A Survey on Multilinguality, Multimodality, Controllability and Learning, Journal of Artificial Intelligence Research, 73:1131-1207, ISSN 1076-9757, April 2022. DOI: 10.1613/jair.1.12918 (Q2 Journal)
2021
2021
Vlad-Iulian Ilie, Ciprian-Octavian Truică, Elena-Simona Apostol, Adrian Paschke. Context-Aware Misinformation Detection: A benchmark of Deep Learning Architectures using Word Embeddings, IEEE Access, 9:162122 - 162146, ISSN: 2169-3536, December 2021 DOI: 10.1109/ACCESS.2021.3132502 (Q2 Journal) [pdf]
Ciprian-Octavian Truică, Elena-Simona Apostol, Maria-Luiza Șerban, Adrian Paschke. Topic-based Document-Level Sentiment Analysis using Contextual Cues, Mathematics, 9(21):1-23(2722), ISSN 2227-7390, October 2021 DOI: 10.3390/math9212722 (Q1 Journal) [pdf]
Software
Software
AWAKEN: https://github.com/elena-apostol/AwakenedCheckThat2022
DocTopic2Vec: https://github.com/cipriantruica/DocTopic2Vec
Context Embeddings: https://github.com/ilievladiulian/misinformation-detection
Page updated
Google Sites
Report abuse