European Commission logo
italiano italiano
CORDIS - Risultati della ricerca dell’UE
CORDIS

WIDER AND ENHANCED VERIFICATION FOR YOU

Risultati finali

WeVerify Open Platform and Benchmark Evaluation v2.0

This software deliverable is the implementation of the WeVerify distributed verification platform and its accompanying software documentation It consists of several key components developed in WPs 4 5 and 6 as well as the content ingestion tools developed in T41 In addition the deliverable will have an accompanying report detailing results from scalability throughput and robustness evaluations Version 20 will be the first integrated prototype of key technologies and tools Scalability of the algorithms from WP5 will be achieved by utilising Storm or a similar streaming framework for continuous processing of new incoming content

Cross-modal verification tools - v2

This will be an update of the verification framework following its integration and testing on top of the platform The update will contain both extensions and new implementations as well as refinements of the previous implementations based on feedback collected from the first evaluation cycles

User-Facing WeVerify Tools v1

This will be the first release of the professional-oriented WeVerify workbench by ATC, coupled with the design and functional specification of the citizen-oriented digital companion by USFD. This will be tested in the first round of pilot activities.

Cross-modal verification tools - v1

his will be the first release of the verification framework, including the definition of the necessary APIs, and first versions of the individual services. The prototype implementation will be accompanied by a report documenting the implemented approaches, their usage, and presenting evaluation results on existing benchmark datasets.

Disinformation Flow Analysis: Tools and Methodology v.2

This will be the updated release of D61 with refined tools and approaches to analysing the flow of disinformation across social networks coupled with a prototype of the early warning system

"Blockchain Database of ""Known Fakes"" v1.0"

This software deliverable will be the result of the implementation of a blockchain distributed ledger that records known fakes The decentralized architecture will be demonstrated by the participation of at least three partners in the blockchain network We will also deliver the methodology for recording transactions in the blockchain through the implementation of appropriate hashing functions for text images and media to be combined with the identity of the verification agent eg a journalist or media organisation A suitable approach to granting user permissions to write in the blockchain will also be implemented according to the type of ledger that will be chosen by the consortium permissioned permissionless publicprivate

User-Facing WeVerify Tools v2

An updated release of D71 with the second fully functional release of the WeVerify workbench by ATC the user interface for searching the decentralised database on known fakes ONTO and the first functional prototype of the digital verification companion USFD This will be available for testing in the second round of pilot activities

WeVerify Open Platform and Benchmark Evaluation v1.0

This software deliverable is the implementation of the WeVerify distributed verification platform and its accompanying software documentation. It consists of several key components, developed in WPs 4, 5, and 6, as well as the content ingestion tools developed in T4.1. In addition, the deliverable will have an accompanying report, detailing results from scalability, throughput, and robustness evaluations. Version 1.0 will be released early in the project and will contain the software tools for collecting content from the sources detailed in T4.1. Since data collection underpins all technical and use case WPs, it is essential to address it early in the project.

WeVerify Open Platform and Benchmark Evaluation v3.0

This software deliverable is the implementation of the WeVerify distributed verification platform and its accompanying software documentation It consists of several key components developed in WPs 4 5 and 6 as well as the content ingestion tools developed in T41 In addition each deliverable will have an accompanying report detailing results from scalability throughput and robustness evaluations Version 30 will integrate further results from WPs 45 and 6 as well as enable crossmedia processing and result storage as well as any changes arising from the use cases WP3 and community pilots WP8 There will be extensive documentation and example code to facilitate takeup

User-Facing WeVerify Tools - Final Version

This deliverable will comprise the final release of the three sets of useroriented WeVerify tools and their documentation following the feedback obtained from the largescale pilot activities

Disinformation Flow Analysis: Tools and Methodology - Final Version

This will be the final release of the Disinformation Flow Analysis framework including crossnetwork analysis and advanced visualisations in order to build a complete view of disinformation flows

Cross-modal verification tools - final version

This will be the final version of the verification framework following extensive evaluation and testing on top of the integrated platform The final version will incorporate improvements and updates based on the results that will have been collected from the last cycles of testing and evaluation

Dissemination Activities Report. v.2

The report will inform about all activities that were carried out in the respective reporting periods

"Blockchain Database of ""Known Fakes"" v2.0"

This software deliverable will be the result of the implementation of a blockchain distributed ledger that records known fakes The decentralized architecture will be demonstrated by the participation of at least three partners in the blockchain network We will also deliver the methodology for recording transactions in the blockchain through the implementation of appropriate hashing functions for text images and media to be combined with the identity of the verification agent eg a journalist or media organisation A suitable approach to granting user permissions to write in the blockchain will also be implemented according to the type of ledger that will be chosen by the consortium permissioned permissionless publicprivate

WeVerify Use Case Analysis Report

This report will describe the different settings where the platform is foreseen to be used and provide concrete user stories and scenarios.

Dissemination Activities Report. v.1

The report will inform about all activities that were carried out in the respective reporting periods

Disinformation Flow Analysis: Tools and Methodology v.1

This will be the first release of the disinformation flow analysis framework to perform network-based disinformation flow analysis, following the work package objectives. We will also define mapping and visualisation objectives of this methodology and relate these to best practice in content verification and misinformation flow analysis.

Final Evaluation Report

This report will present the outcomes of the second phase evaluation activities also providing an overview of progress achieved during the whole project

WeVerify Architecture Definition

This report will define the architecture and its APIs (Application Programming Interfaces), design decisions, evaluation and unit testing, release schedule, and deployment targets.

Project Website and Initial Set of Promotional Materials

Launch of the project website and availability of first set of promotional materials.

Pubblicazioni

Team Bertha von Suttner at SemEval-2019 Task 4: Hyperpartisan News Detection using ELMo Sentence Representation Convolutional Network

Autori: Ye Jiang, Johann Petrak, Xingyi Song, Kalina Bontcheva, Diana Maynard
Pubblicato in: Proceedings of the 13th International Workshop on Semantic Evaluation, 2019, Pagina/e 840-844
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/s19-2146

Adapting text mining tools to noisy text

Autori: Maynard, Diana
Pubblicato in: Invited talk given at Text Mining for Science Studies Workshop, Numero 4, 2019
Editore: Zenodo
DOI: 10.5281/zenodo.3609868

Audio-based Near-Duplicate Video Retrieval with Audio Similarity Learning

Autori: Pavlos Avgoustinakis; Giorgos Kordopatis-Zilos; Symeon Papadopoulos; Andreas L. Symeonidis; Ioannis Kompatsiaris
Pubblicato in: ICPR, Numero 2, 2020, Pagina/e 5828-5835
Editore: IEEE
DOI: 10.1109/icpr48806.2021.9413056

Measuring What Counts: The case of Rumour Stance Classification

Autori: Scarton, Carolina; Silva, Diego F.; Bontcheva, Kalina
Pubblicato in: Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing, 2020, Pagina/e 925-932
Editore: Association for Computational Linguistics

Adding value to NLP: a little semantics goes a long way

Autori: Maynard, Diana
Pubblicato in: Keynote talk presented at ESWC 2019, 2019
Editore: Zenodo
DOI: 10.5281/zenodo.3609847

Context Aggregation and Analysis: A Tool for User-Generated Video Verification

Autori: Olga Papadopoulou; Dimitrios Giomelakis; Lazaros Apostolidis; Symeon Papadopoulos; Ioannis Kompatsiaris
Pubblicato in: Proceedings of the Workshop on Reducing Online Misinformation Exposure ROME 2019, 2019
Editore: ROME2019

Knowledge-Based Fusion for Image Tampering Localization

Autori: Chryssanthi IakovidouSymeon Papadopoulos, Yiannis Kompatsiaris
Pubblicato in: IFIP Advances in Information and Communication Technology, Numero vol. 583, 2020, Pagina/e 177-188, ISBN 978-3-030-49161-1
Editore: Springer, Cham
DOI: 10.1007/978-3-030-49161-1_16

WeVerify: Wider and Enhanced Verification for You Project Overview and Tools

Autori: Zlatina Marinova, Jochen Spangenberg, Denis Teyssou, Symeon Papadopoulos, Nikos Sarris, Alexandre Alaphilippe, Kalina Bontcheva
Pubblicato in: IEEE International Conference on Multimedia and Expo Workshops (ICMEW), 2020, ISBN 978-1-7281-1485-9
Editore: IEEE
DOI: 10.1109/icmew46912.2020.9106056

Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation

Autori: Giorgos Kordopatis-Zilos; Panagiotis Galopoulos; Symeon Papadopoulos; Ioannis Kompatsiaris
Pubblicato in: ICMR, Numero 1, 2021, Pagina/e 155-163
Editore: ACM
DOI: 10.1145/3460426.3463644

Journalist-in-the-Loop: Continuous Learning as a Service for Rumour Analysis

Autori: Twin Karmakharm, Nikolaos Aletras, Kalina Bontcheva
Pubblicato in: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations, 2019, Pagina/e 115-120
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/d19-3020

Cross-Lingual Rumour Stance Classification: a First Study with BERT and Machine Translation

Autori: Carolina Scarton, Yue Li
Pubblicato in: Proceedings of the 2021 Truth and Trust Online Conference (TTO 2021), 2021, Pagina/e 50-59, ISBN 978-1-7359904-0-8
Editore: Hacks Hackers

Tackling Multilinguality and Internationality in Fake News

Autori: Andrey Tagarev, Krasimira Bozhanova, Ivelina Nikolova-Koleva, Ivan Ivanov
Pubblicato in: Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2021), 2021, Pagina/e 1380-1386
Editore: INCOMA Ltd.

MeVer team tackling Coronavirus and 5G conspiracy using ensemble classification based on BERT

Autori: Olga Papadopoulou, Giorgos Kordopatis-Zilos, Symeon Papadopoulos
Pubblicato in: Working Notes Proceedings of the MediaEval 2020 Workshop, 2020
Editore: CEUR workshop proceedings

ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning

Autori: Kordopatis-Zilos, Giorgos; Papadopoulos, Symeon; Patras, Ioannis; Kompatsiaris, Ioannis
Pubblicato in: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, Pagina/e 6351-6360
Editore: IEEE/CVF
DOI: 10.5281/zenodo.3615769

Revisiting Rumour Stance Classification: Dealing with Imbalanced Data

Autori: Yue Li, Carolina Scarton
Pubblicato in: Proceedings of the 3rd International Workshop on Rumours and Deception in Social Media (RDSM), 2020
Editore: Association for Computational Linguistics

Leveraging Selective Prediction for Reliable Image Geolocation

Autori: Panagiotopoulos, A., Kordopatis-Zilos, G., & Papadopoulos, S.
Pubblicato in: 28th International Conference on Multimedia Modeling, 2021
Editore: ArXiv.org
DOI: 10.48550/arxiv.2111.11952

Brenda Starr at SemEval-2019 Task 4: Hyperpartisan News Detection

Autori: Olga Papadopoulou, Giorgos Kordopatis-Zilos, Markos Zampoglou, Symeon Papadopoulos, Yiannis Kompatsiaris
Pubblicato in: Proceedings of the 13th International Workshop on Semantic Evaluation, 2019, Pagina/e 924-928
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/s19-2157

Investigating the Impact of Pre-processing and Prediction Aggregation on the DeepFake Detection Task

Autori: Charitidis, Polychronis; Kordopatis-Zilos, Giorgos; Papadopoulos, Symeon; Kompatsiaris, Ioannis
Pubblicato in: Proceedings of the 2020 Truth and Trust Online Conference (TTO 2020), Numero 1, 2020, Pagina/e 44-54, ISBN 978-1-7359904-0-8
Editore: Hacks Hackers 2020
DOI: 10.48550/arxiv.2006.07084

Operation-wise Attention Network for Tampering Localization Fusion

Autori: Charitidis, P., Kordopatis-Zilos, G., Papadopoulos, S., & Kompatsiaris, I.
Pubblicato in: 2021 International Conference on Content-Based Multimedia Indexing (CBMI), 2021, Pagina/e 1-6, ISBN 978-1-6654-4220-6
Editore: IEEE
DOI: 10.1109/cbmi50038.2021.9461920

Multistage BiCross encoder for multilingual access to COVID-19 health information.

Autori: Iknoor Singh; Carolina Scarton; Kalina Bontcheva
Pubblicato in: PLOS ONE, Numero 2, 2021, ISSN 1932-6203
Editore: Public Library of Science
DOI: 10.1371/journal.pone.0256874

FIVR: Fine-Grained Incident Video Retrieval

Autori: Giorgos Kordopatis-Zilos, Symeon Papadopoulos, Ioannis Patras, Ioannis Kompatsiaris
Pubblicato in: IEEE Transactions on Multimedia, Numero 21/10, 2019, Pagina/e 2638-2652, ISSN 1520-9210
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/TMM.2019.2905741

Classification aware neural topic model for COVID-19 disinformation categorisation.

Autori: Xingyi Song; Johann Petrak; Johann Petrak; Ye Jiang; Iknoor Singh; Iknoor Singh; Diana Maynard; Kalina Bontcheva
Pubblicato in: PLoS ONE, Numero 4, 2021, ISSN 1932-6203
Editore: Public Library of Science
DOI: 10.1371/journal.pone.0247086

Online Abuse of UK MPs from 2015 to 2019: Working Paper

Autori: Greenwood, M.A.; Bakir, M.E.; Gorrell, G.; Song, X.; Roberts, I.; Bontcheva, K.
Pubblicato in: arXiv, Numero 2, 2019
Editore: University of Sheffield
DOI: 10.5281/zenodo.4081280

Race and Religion in Online Abuse towards UK Politicians: Working Paper

Autori: Gorrell, G.; Bakir, M.E.; Greenwood, M.A.; Roberts, I.; Bontcheva, K.
Pubblicato in: arXiv, Numero 2, 2019
Editore: University of Sheffield
DOI: 10.5281/zenodo.4081270

Categorising fine-to-coarse grained misinformation : an empirical study of COVID-19 infodemic

Autori: Jiang, Y.; Song, X.; Scarton, C.; Aker, A.; Bontcheva, K.
Pubblicato in: arXiv, Numero 2, 2021, ISSN 2331-8422
Editore: ArXiv.org

The false COVID-19 narratives that keep being debunked : a spatiotemporal analysis

Autori: Singh, I.; Bontcheva, K.; Scarton, C.
Pubblicato in: arXiv, Numero 2, 2021
Editore: White Rose Research Online
DOI: 10.48550/arxiv.2107.12303

DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval

Autori: Kordopatis-Zilos, Giorgos; Tzelepis, Christos; Papadopoulos, Symeon; Kompatsiaris, Ioannis; Patras, Ioannis
Pubblicato in: Numero 3, 2021
Editore: ArXiv.org
DOI: 10.48550/arxiv.2106.13266

Verification of News Video Content: Findings from a Study of Journalism Students

Autori: Dimitrios Giomelakis; Olga Papadopoulou; Symeon Papadopoulos; Andreas Veglis
Pubblicato in: Journalism Practice, Numero 1, 2021, ISSN 1751-2794
Editore: Taylor & Francis Online
DOI: 10.1080/17512786.2021.1965905

È in corso la ricerca di dati su OpenAIRE...

Si è verificato un errore durante la ricerca dei dati su OpenAIRE

Nessun risultato disponibile