AI-Based Harassment Detector for Social Media

This project implements an AI-based machine learning model designed to detect and classify misogynistic, abusive, and harmful comments directed at women on social media platforms. The system identifies inappropriate content in real-time, enabling further moderation actions. Built using traditional machine learning techniques, the model focuses on accurate classification to support safer online environments.

---

Table of Contents

1. Installation
2. Setup Instructions
3. Running the Model

---

Installation

To get this project up and running, you’ll need Python installed on your system. We’ll also be using a virtual environment to keep all the dependencies organized (trust me, it’s way cleaner that way!). Just follow the steps below to get everything set up and ready for action.

---

Setup Instructions

Follow these steps to get the project running locally on your machine:

1. Clone the Repository

First, clone the project to your local machine with this command:

git clone https://github.com/your-username/your-repo.git

2. Navigate to the Project Directory

Next, head into the project directory:

cd your-repo

3. Set Up a Virtual Environment

Create a fresh virtual environment to keep your project dependencies clean and isolated:

python -m venv venv

4. Activate the Virtual Environment

Now, activate the virtual environment to make sure all the dependencies get installed in the right place:

• On Windows:

  venv\Scripts\activate

• On MacOS:

  source venv/bin/activate

5. Install Dependencies

With your virtual environment up and running, install all the necessary dependencies from the requirements.txt file:

pip install -r requirements.txt

---

Running the Model

Once everything is set up, you're ready to run the model:

• To train the model and understand how it works, run this command:

  python train_model.py

• To test the model and check out its performance, execute:

  python test_model.py

Muslimah Sarumi | University of Bolton | 2025