https://github.com/bhavikagarg/hate-speech-offensive-language
Addresses the critical issue of hate speech and offensive language proliferating on online platforms.
https://github.com/bhavikagarg/hate-speech-offensive-language
deep-learning hate-speech-detection lstm-neural-networks machine-learning python sentiment-analysis
Last synced: about 2 months ago
JSON representation
Addresses the critical issue of hate speech and offensive language proliferating on online platforms.
- Host: GitHub
- URL: https://github.com/bhavikagarg/hate-speech-offensive-language
- Owner: bhavikagarg
- Created: 2025-03-22T15:12:22.000Z (about 2 months ago)
- Default Branch: main
- Last Pushed: 2025-03-22T15:30:54.000Z (about 2 months ago)
- Last Synced: 2025-03-22T16:30:39.797Z (about 2 months ago)
- Topics: deep-learning, hate-speech-detection, lstm-neural-networks, machine-learning, python, sentiment-analysis
- Homepage: https://www.kaggle.com/datasets/mrmorj/hate-speech-and-offensive-language-dataset
- Size: 1.11 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Hate-Speech-Offensive-Language
# 📖 Overview
**Hate speech and offensive language detection** is a crucial task in **natural language processing (NLP)** to identify and mitigate harmful content online. This project aims to develop a **deep learning model** using **Long Short-Term Memory (LSTM)** to classify text into three categories:
✅ **Hate Speech**
✅ **Offensive Language**
✅ **Neutral (Neither Hate nor Offensive)**
To ensure **fairness, transparency, and interpretability**, we integrate **Explainable AI (XAI) techniques** like **SHAP and LIME** to understand the model’s decision-making process.
---
## 🎯 **Problem Statement**
With the rise of **social media platforms**, monitoring and regulating hate speech is challenging. Existing methods either **lack accuracy, struggle with contextual understanding, or fail to provide interpretability**. The key challenges include:
- **Complexity of language** – Hate speech can be implicit or subtle.
- **Data Imbalance** – Some classes (e.g., Offensive Language) have fewer samples.
- **Contextual Misinterpretation** – Some words may seem offensive but are not.
- **Lack of Transparency** – Many AI models work as “black boxes.”
---
## 📂 **Dataset Information**
The dataset used consists of **tweets labeled into three categories**:
| **Class** | **Count** |
|--------------------|----------|
| Hate Speech | 19,190 |
| Offensive Language | 1,430 |
| Neither | 4,163 |
**1️⃣ Clone the Repository**
```sh
git clone https://github.com/your-username/hate-speech-detection.git
cd hate-speech-detection