- Image classification using Deep learning. - Utilizing both frequency and pixel domain information of images. - Implemented MVNN model from a research paper published in 2019 IEEE ICDM. Achieved comparable accuracy to the original model. - Compared with several deep learning models: Pre-trained (VGG-16, VGG-19) and our own implementations of CNN based models (with different number of layers) - Research project utilized as a part of a course: COMP 5331 - Knowledge Discovery in Databases, from HKUST (The Hong Kong University of Science and Technology). - Referenced paper (for MVNN model) - P. Qi, J. Cao, T. Yang, J. Guo, and J. Li. Exploiting multi-domain visual information for fake news detection. In 2019 IEEE International Conference on Data Mining (ICDM), pages 518–527, 2019.
