Emoji Predictor with BERTweet

Live Demo

Description

This project is focused on the relationship between text data and emojis. Using a fine-tuned version of the state-of-the-art model BERTweet, we've managed to improve top-1 accuracy by 10%. The dataset used for this study was developed uniquely for this task by scraping tweets and prompting the GPT 3.5 Turbo API for emoji prediction.

The application is built using Next.js for the frontend and Flask for the backend. It maps Unicode emojis to Microsoft's animated Fluent Emojis.

Getting Started

Follow these instructions to get a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

Node.js (Download and Install from here)
Python 3 (Download and Install from here)
Flask (Python Web Framework)
Pip (Python Package Installer)
Download my fine tuned BERTweet model from here and place it in the backend directory.

Installation

Clone the repository:

https://github.com/itsEricWu/emoji-predictor-web.git
cd emoji-predictor-web

Install Flask and the necessary Python packages using pip:
```
pip install -r requirements.txt
```
Navigate to the frontend directory:
```
cd frontend
```
Install the required Node packages:
```
npm install
```

Running the Application

Start the Flask server: Navigate to the backend directory and run:
```
cd ../backend
python3 app.py
```
The server should now be running on http://localhost:5000.
Start the Next.js application: In a new terminal, navigate to the frontend directory and run:
```
cd frontend
npm run dev
```
The application should now be running on http://localhost:3000.