Cluster-based deep ensemble learning for emotion classification in Internet memes

Memes have gained popularity as a means to share visual ideas through the Internet and social media by mixing text, images videos, often for humorous purposes. Research enabling automated analysis of memes has attention in recent years, including among others task classifying emotion expressed memes. In this article, we propose novel model, cluster-based deep ensemble learning (CDEL), classification CDEL is hybrid model that leverages benefits combination with clustering algorithm, which enhances additional information after similar facial features. We evaluate performance on benchmark data set classification, proving its effectiveness outperforming wide range baseline models achieving state-of-the-art performance. Further evaluation ablated demonstrates different components CDEL.