akshat14714 / correlating-text-with-emoji

SMAI Course Project

Correlating-Text-with-Emoji

SMAI semster course project

Team Members

• Kalpit Pokra (20161134)
• Kushagra Nagori (20161032)
• Akshat Maheshwari (20161024)
• Kartik Garg (20161219)

Goal

The goal of the project is: Given a tweet, predict the most likely emoji that was used along with it.

Variants

• Classification : Given a tweet, predict the most likely emoji that was used along with it.

• Emoji Embeddings : Given a tweet without an emoji, find its embedding (averaging of word embeddings), and find the closest emoji to it. This can be extended to: given a statement, get the most relevant emoji that is suitable with that sentence using semantic analysis of that sentence using the word2vec and the bag-of-words models. This can then be extended to paragraphs. Given a paragraph, break the paragraph to sentences and analyze each sentence and find the emoji associated with that sentence using the previous model.

Most General Models

The most general models that we can use for the classification tasks are LinearSVC, Naive Bayes, Logistic Regression, KNN Classifier, and other classification models.

Proposal

This is the link for the doc.