Abstract
The growth of multilingual web content in lowresource languages is becoming an emergingchallenge to detect misinformation. One particular hindrance to research on this problemis the non-availability of resources and tools.Majority of the earlier works in misinformation detection are based on English contentwhich confines the applicability of the researchto a specific language only. Increasing presence of multimedia content on the web haspromoted misinformation in which real multimedia content (images, videos) are used indifferent but related contexts with manipulatedtexts to mislead the readers. Detecting thiscategory of misleading information is almostimpossible without any prior knowledge. Studies say that emotion-invoking and highly novelcontent accelerates the dissemination of falseinformation. To counter this problem, here inthis paper, we first introduce a novel multilingual multimodal misinformation dataset thatincludes background knowledge (from authentic sources) of the misleading articles. Second,we propose an effective neural model leveraging novelty detection and emotion recognitionto detect fabricated information. We performextensive experiments to justify that our proposed model outperforms the state-of-the-art(SOTA) on the concerned task1.
Original language | English |
---|---|
Title of host publication | 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing - Findings of the Association for Computational Linguistics |
Subtitle of host publication | AACL-IJCNLP 2022 |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 464-477 |
Number of pages | 14 |
ISBN (Electronic) | 9781959429043 |
State | Published - 2022 |
Externally published | Yes |
Event | 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing, AACL-IJCNLP 2022 - Virtual, Online, Taiwan, Province of China Duration: Nov 20 2022 → Nov 23 2022 |
Publication series
Name | 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing - Findings of the Association for Computational Linguistics: AACL-IJCNLP 2022 |
---|
Conference
Conference | 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing, AACL-IJCNLP 2022 |
---|---|
Country/Territory | Taiwan, Province of China |
City | Virtual, Online |
Period | 11/20/22 → 11/23/22 |
Funding
This reserach is partially supported by the project “HELIOS - Hate, Hyperpartisan, and Hyperplural-ism Elicitation and Observer System“, sponsored by Wipro. We thank all the anonymous reviewers for their insightful suggestions. We also sincerely thank Tanik Saikh for his contribution in data annotation of Bengali Language. All the research work was carried out at AI-ML-NLP research lab, IIT Patna.