New Modality: Emoji Challenges in Prediction, Anticipation, and Retrieval
To predict and Anticipation Emoji for multimedia retrieval.
To facilitate further research on emoji, we propose three emoji challenge problems and present state-of-the- art neural network baselines for them, as well as a dataset for evaluation. Previously cannot find a Emoji correctly and not accuracy. To found a unrelated emoji are displayed. So, we will move to the proposed model.
We explore the challenges that arise naturally from considering the emoji modality through the lens of multimedia research, specifically the ways in which emoji can be related to other common modalities such as text and images. To do so, we first present a large-scale data set of real-world emoji usage collected from Twitter. This data set contains examples of both text emoji and image-emoji relationships within tweets. We present baseline results on the challenge of predicting emoji from both text and images, using state-of-the-art neural networks.