Tools
Tools
Waghela, Prashant Digambar (2022) Multi-Modal Fake News and Tampered Image Detection using Transformer and CNN-based Algorithms. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (426kB) | Preview |
Abstract
In this modern world, the use of computing devices and internet access has made it easier for any news article or post to spread among the masses, resulting in quick access to information but, sometimes these platforms are exploited to spread fake information. A customized multimodal algorithm that uses these news headlines to verify the authenticity and concurrently recognize if its corresponding image information is fabricated or not can reduce the spread of wrong information. The report provides a thorough literature review that helped in devising this multimodal technique for fake news classification. Moreover, the research contributes by developing three unique multimodal algorithms BERT+CNN, BERT+InceptionV3, and XML_RoBERTa+CNN that classify fake news text and related images simultaneously. The multimodal BERT+CNN model provided the best accuracy of 71% which was comparable to the unimodal BERT approach which achieved 72% accuracy. The study was crucial in understanding the impact of using multimodal text and visual features to classify fake news and the obtained results were analyzed to extract insights from the implemented multimodal technique.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | BERT+CNN; BERT+InceptionV3; XML_RoBERTa+CNN |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 14 Mar 2023 14:31 |
| Last Modified: | 14 Mar 2023 14:31 |
| URI: | https://norma.ncirl.ie/id/eprint/6339 |
Actions (login required)
 |
View Item |