🧠 PSO for AI Hyperparameter Tuning & Predictive Analytics 🚀

📌 Project Overview

This project applies Particle Swarm Optimization (PSO) to optimize Support Vector Machine (SVM) hyperparameters, demonstrating the power of Swarm Intelligence in AI.

✅ Optimized AI models without brute-force search
✅ Achieved 100% accuracy on the Iris dataset
✅ Showcased decentralized decision-making in AI

🔬 Methodology

• Dataset: Iris classification dataset 🌸
• Optimization Target: SVM hyperparameters (C, Gamma)
• Search Range:
  • C: [0.1, 100]
  • Gamma: [0.0001, 1]
• Swarm Details: 20 particles, 30 iterations

📊 Results

• Optimal Parameters Found:
  • C = 12.8346
  • Gamma = 0.6106
• Achieved Accuracy: 100%

📂 Files in this Repository

• swarm_ai_pso.py → Python implementation of PSO
• PSO_Hyperparameter_Tuning_Report.pdf → Detailed experiment report
• Enhanced_PSO_AI_Presentation_Light_Background.pptx → Presentation slides

🚀 Next Steps

• Apply PSO for time-series forecasting 📈
• Extend to multi-agent AI models
• Optimize deep learning architectures using Swarm Intelligence

---

📌 Follow & Contribute: [Your GitHub Profile]
🔗 Read More About Swarm Intelligence: [Link to article]