📄 Before It's Too Late: A State Space Model for the Early Prediction of Misinformation and Disinformation Engagement

📝 Introduction

In today’s digital age, misinformation and disinformation can rapidly spread, impacting social and democratic cohesion. While deep learning models have made strides in engagement prediction, they struggle with irregularly sampled data and early trajectory forecasting.

We introduce IC-Mamba, a novel state space model that forecasts social media engagement by integrating interval-censored data with temporal embeddings. Our model is particularly effective in predicting engagement patterns within the crucial first 15-30 minutes of posting (RMSE 0.118-0.143), offering early insights into content virality.

🔹 Key Contributions:

• A state space framework for modeling engagement with interval-censored data.
• A 4.72% improvement over state-of-the-art engagement prediction models.
• Long-term engagement forecasting: up to 28 days ahead based on 3-10 days of observations.
• Enables earlier identification of potentially problematic content for proactive countermeasures.

---

📊 Interactive Dashboard

Explore the live IC-Mamba results through our interactive dashboard:

➡️ IC-Mamba Dashboard

---

🏗 Installation

To set up the environment and run IC-Mamba, use:

git clone https://github.com/ltian678/ic-mamba.git
cd ic-mamba
pip install -r requirements.txt

📩 Contact

For any questions or collaborations, feel free to reach out:

• Lin Tian - lin.tian-3@uts.edu.au
• Marian-Andrei Rizoiu - marian-andrei.rizoiu@uts.edu.au

---

🏷 License

This project is licensed under the Creative Commons Attribution 4.0 International License.

You are free to share, adapt, and build upon this work with attribution. Please refer to the full license here:
🔗 Creative Commons Attribution 4.0 International License

---

🌟 Star the repository if you find this useful!