BYRD's I - Twitter Sentiment Analysis Project

Overview

BYRD's I is a collaborative project developed as part of a team at the University at Albany, SUNY, that aims to analyze and classify the sentiment behind text data. The project focuses on Twitter data and uses a machine learning model to determine whether the sentiment of tweets is positive, negative, or neutral.

We trained the sentiment analysis model using over 8 different datasets to ensure its reliability and accuracy. This project was also made open-source for the wider community to use and has been downloaded more than 90 times since its release.

Key Features

Sentiment Analysis Model:
- Analyzes text and classifies it as positive, negative, or neutral.
- Trained using multiple datasets for high accuracy in sentiment detection.
Application to Twitter Data:
- Specifically applied to the sentiment analysis of celebrity tweets, demonstrating its effectiveness in real-world data.
Open-Source Release:
- Published the model for public use, with significant attention from the developer and research community.

Tech Stack

Machine Learning Model: Hugging Face’s roberta-base-tweet-eval model
Frontend: Node.js, JavaScript
Backend: REST APIs
Deployment: AWS
Version Control: Git/GitHub
Project Management: Waterfall Methodology

Development & Setup

This project demonstrates proficiency in machine learning, data preprocessing, and model training. The model is available on Hugging Face for open-source usage:

Install dependencies:

git clone https://github.com/Caffeine-Coders/Sentiment-Analysis-Project.git
cd byrds-i
npm install

Run the application:
```
npm start
```

Access & Documentation

This project highlights my expertise in machine learning, full-stack development, and deploying scalable solutions on cloud platforms like AWS.