NetGuardAI - MultiPRIDE 2026

This repository contains the source code and experimental logs for Team NetGuardAI's submission to the EVALITA 2026 MultiPRIDE Shared Task.

📂 Files

• NetGuardAI_MultiPRIDE_Submission.ipynb: The complete Jupyter Notebook containing:
  • Exploratory Data Analysis (Statistics & Topic Modeling)
  • Data Preprocessing & Tokenization
  • DistilBERT Model Training (Task A & Task B)
  • Weighted Loss Implementation
  • Evaluation & Error Analysis

🚀 How to Run

The code was designed to run in Google Colab.

1. Open the .ipynb file in Google Colab or Jupyter Lab.
2. Install dependencies:

   pip install transformers torch scikit-learn pandas numpy