Emoji Prediction Using Bi-Directional LSTM

Messengers and social media dominate today’s internet usage across the globe. For large population, a typical day starts with messages flooding on mobiles, from simple good morning wishes, business meeting invites, reminders, schedules for list is endless. A striking feature of digital communication variety emojis used, without which text almost look incomplete. Emojis are graphic symbols/logograms used to enhance effectiveness emotions set an undertone that makes texting more fun experience users. visual language new generation. They give consumers means communicate their feelings while reducing quantity needs be typed by sender. Every messenger platform like Facebook, Instagram, Twitter, WhatsApp, many have its own emoji set. To lure users, added day. Predicting suggesting based text, emotion user patterns important messengers applications. If you start typing message, relevant will displayed users can choose emoji, further enhancing experience. This process done using natural processing machine learning techniques. In this paper, we study prediction techniques propose model bi-directional LSTMs. We compare NLP techniques, including RNN, LSTM, LSTM networks, Bi-LSTM. Based our implementation, suggest most effective technique. Our outperforms baseline approaches accuracy 94% when tested CodaLab Twitter data 60000 rows two columns. shows efficiency LSTMs text-based systems communication.