Diabetes Prediction Model

A machine learning-based application that predicts whether a person is likely to have diabetes based on medical input parameters. The project uses a trained Support Vector Machine (SVM) model and provides a simple Streamlit interface for user interaction.

---

🚀 Features

• Predicts diabetes risk using medical data
• Machine Learning model built with SVM
• Simple and interactive Streamlit web interface
• Fast predictions with a pre-trained model
• Easy to run locally

---

🧠 Tech Stack

• Python
• NumPy
• Pandas
• Scikit-learn
• Streamlit

---

📂 Project Structure

Diabetes_Prediction_Model
│
├── diabetes.csv                 # Dataset
├── diabetes.py                  # Streamlit app
├── diabetes_model_new.sav       # Trained ML model
├── Streamlit interface.txt      # Streamlit run instructions
└── torun.txt                    # Execution instructions

---

📊 Dataset

The dataset contains medical attributes such as:

• Pregnancies
• Glucose level
• Blood pressure
• Skin thickness
• Insulin
• BMI
• Diabetes pedigree function
• Age

These features are used to predict whether a person is diabetic or not.

---

⚙️ Installation & Setup

1. Clone the repository

git clone https://github.com/your-username/Diabetes_Prediction_Model.git
cd Diabetes_Prediction_Model

2. Install dependencies

pip install numpy pandas scikit-learn streamlit

3. Run the application

streamlit run diabetes.py

---

🖥️ How It Works

1. User enters medical parameters in the Streamlit interface.

2. The input data is passed to the trained SVM model.

3. The model predicts whether the person is:

   • Diabetic
   • Non-diabetic

4. Result is displayed instantly on the web interface.

---

📌 Future Improvements

• Deploy the application online
• Improve model accuracy with advanced algorithms
• Add user authentication
• Store prediction history

---

This project is for educational purposes.