AI-Driven Process Recommendation System 🚀

An advanced, hybrid process monitoring and recommendation system that combines deterministic Rule-Based logic with Machine Learning (Random Forest) to optimize system performance and user workflow.

---

🌟 Overview

This project serves as an intelligent process manager for modern operating systems. It doesn't just list processes; it analyzes them. By monitoring CPU, memory, and thread usage in real-time, the system provides actionable recommendations (Prioritize, Monitor, Pause, Terminate) to keep your workspace smooth and efficient.

The Hybrid Advantage

• v1.0 (Rule Engine): Reliable, transparent logic based on resource thresholds and process categories (System, Browser, Dev Tools, etc.).
• v2.0 (ML Model): A Random Forest Classifier that learns from your behavior. As you label processes as "Critical" or "Expendable", the AI adapts its recommendations to match your specific needs.

---

🛠️ Tech Stack

• Backend: Python, Flask, psutil (System Monitoring), scikit-learn (Machine Learning).
• Frontend: React (JSX), Vanilla CSS (Modern Glassmorphism UI).
• Data Storage: JSON for metadata, Pickle for model serialization.
• Algorithms: Random Forest Classifier, Rule-Based Heuristics.

---

🚀 Key Features

1. Intelligent Monitoring

• Real-time Stats: CPU Usage (per core), RAM consumption, Swap memory, and CPU Temperature.
• Process Categorization: Automatically detects if a process is a System tool, Browser, Media app, or Developer tool.
• Resource Composite Score: Sorts processes by a weighted balance of CPU (60%) and Memory (40%) to highlight the most impactful tasks.

2. User-Driven AI Training

• Manual Labeling: Users can explicitly label processes.
• Auto-Training: The system automatically re-trains the ML model as you provide feedback, improving its accuracy over time.
• Model Insights: View feature importance (e.g., how much "Memory %" influences the AI's decision).

3. Dynamic Process Control

• Tree-Aware Actions: Killing or pausing a process automatically handles all its child processes recursively.
• Priority Boosting: Use the "Prioritize" action to give critical tools more CPU cycles.

---

📂 Project Structure

• process_backend.py: The unified API server and core logic engine.
• process_dashboard.jsx: The premium React frontend (served directly by the backend).
• model_metadata.json: Real-time training metrics and feature importances.
• process_model.pkl: The saved state of your personalized AI model.

---

⚙️ Installation & Setup

1. Prerequisites

Ensure you have Python 3.8+ installed.

2. Install Dependencies

pip install flask flask-cors psutil scikit-learn numpy

3. Run the Application

python process_backend.py

After running, open your browser to: http://localhost:5000

---

📖 How to Use

1. Monitor: Use the dashboard to see which processes are consuming the most resources.
2. Label: Help the AI learn! Label your favorite tools as Critical and background junk as Expendable.
3. Train: Click the "Train" button (if triggered or after enough labels) to switch the engine from Rule-Based to ML-Driven.
4. Action: Use the action buttons (Kill, Pause, Resume, Boost) to manage your system directly from the web interface.

---

🛡️ Safety Note

The system includes a System Process Protection layer. Processes categorized as "System" (like kernel, lsass, or svchost) are automatically prioritized and protected from accidental termination by the rule engine.

---

Created as part of the Process-Scheduler-OS project.