https://github.com/jibbs1703/classify-text-models

This repository contains projects that classify texts using a variety of machine learning and deep learning models. The projects show use-cases of classifying text data through Natural Language Processing methods.
https://github.com/jibbs1703/classify-text-models

data-science deep-learning machine-learning natural-language-processing text-classification

Last synced: over 1 year ago
JSON representation

This repository contains projects that classify texts using a variety of machine learning and deep learning models. The projects show use-cases of classifying text data through Natural Language Processing methods.

• Host: GitHub
• URL: https://github.com/jibbs1703/classify-text-models
• Owner: jibbs1703
• License: mit
• Created: 2024-04-02T12:35:51.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2025-02-02T21:54:50.000Z (over 1 year ago)
• Last Synced: 2025-02-02T22:28:07.709Z (over 1 year ago)
• Topics: data-science, deep-learning, machine-learning, natural-language-processing, text-classification
• Language: Python
• Homepage:
• Size: 680 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Text Classification 

## Overview

This repository contains projects that classify text data using a variety of machine learning and deep learning models. The projects demonstrate real-world use-cases of Natural Language Processing. The tasks completed cover disaster tweet classification, spam message detection, and fake news recognition.

## Projects

### **Disaster Tweets Classification**

- **Dataset**: Disaster tweets from [Kaggle](https://www.kaggle.com/competitions/nlp-getting-started).

- **Achievement**: F1 accuracy scores: Logistic Regression (79.25%), Complement Naive Bayes (78.42%).

- **Process**: Developed supervised machine learning models to classify tweets as disaster-related (1) or non-disaster-related (0).

### **Fake News Recognition**

- **Dataset**: Fake and real news articles from [Kaggle](https://www.kaggle.com/datasets/clmentbisaillon/fake-and-real-news-dataset/data).

- **Achievement**: Logistic Regression Cross Validation Model achieved 99.51% F1 score.

- **Process**: Developed models to classify news articles as true (1) or fake (0).

### **Spam Message Detection**

- **Dataset**: Spam email messages from [Kaggle](https://www.kaggle.com/datasets/ashfakyeafi/spam-email-classification).

- **Achievement**: Logistic Regression Cross Validation Model achieved 94.37% F1 score.

- **Process**: Developed models to classify emails as spam (1) or ham (0).