🤖 AutoML + AI Insights Platform

Build • Compare • Analyze • Improve Machine Learning Models with Google Gemini AI

An AI-powered AutoML application that automatically preprocesses datasets, detects the machine learning problem type, trains multiple models, compares their performance, and generates intelligent model analysis and improvement suggestions using Google Gemini AI.

---

🚀 Project Overview

Selecting the right Machine Learning model and interpreting its performance can be challenging, especially for beginners. This application automates the entire workflow by combining traditional Machine Learning with Generative AI.

Users simply upload a CSV dataset, choose the target column, and the application handles preprocessing, feature engineering, model training, evaluation, visualization, and AI-generated insights automatically.

---

✨ Key Features

• 📂 Upload any CSV dataset
• 🧹 Automatic data preprocessing
• 🔍 Automatic problem type detection (Classification or Regression)
• 🔠 Missing value handling
• 🔢 Categorical encoding
• 📏 Feature scaling
• 🤖 Automatic model training
• 📊 Model performance comparison
• 📈 Interactive visualizations
• 🧠 AI-generated performance summaries
• 💡 AI-powered model improvement suggestions
• 📥 Download model results as CSV

---

🛠 Technology Stack

Programming

• Python

Machine Learning

• Scikit-learn
• Logistic Regression
• Linear Regression
• Random Forest
• Gradient Boosting

AI

• Google Gemini 2.5 Flash API

Web Framework

• Streamlit

Data Processing

• Pandas
• NumPy

Visualization

• Streamlit Charts

Environment

• python-dotenv

---

⚙️ Workflow

CSV Dataset
      │
      ▼
Upload Dataset
      │
      ▼
Automatic Data Preprocessing
      │
      ▼
Missing Value Handling
      │
      ▼
Categorical Encoding
      │
      ▼
Feature Scaling
      │
      ▼
Problem Detection
(Classification / Regression)
      │
      ▼
Multiple Model Training
      │
      ▼
Performance Evaluation
      │
      ▼
Interactive Charts
      │
      ▼
Google Gemini AI
      │
      ▼
Performance Summary
      │
      ▼
Improvement Suggestions

---

🤖 Machine Learning Models

Classification

• Logistic Regression
• Random Forest Classifier
• Gradient Boosting Classifier

Regression

• Linear Regression
• Random Forest Regressor
• Gradient Boosting Regressor

---

📊 Evaluation Metrics

Classification

• Accuracy
• Precision
• Recall
• F1 Score

Regression

• R² Score
• RMSE (Root Mean Squared Error)

---

🧠 AI-Powered Insights

After model evaluation, the application uses Google Gemini 2.5 Flash to generate:

📑 Performance Summary

• Identifies the best-performing model
• Explains why it performs best
• Summarizes model performance

💡 Improvement Suggestions

• Feature engineering recommendations
• Better algorithm suggestions
• Data preprocessing improvements
• General model optimization tips

---

📂 Project Structure

AutoML-AI-Insights
│
├── app.py
├── analysis.py
├── requirements.txt
├── .env
├── README.md
└── sample_dataset.csv

---

🚀 Installation

Clone the repository

git clone https://github.com/yourusername/AutoML-AI-Insights.git

Move into the project directory

cd AutoML-AI-Insights

Install dependencies

pip install -r requirements.txt

Create a .env file

GOOGLE_API_KEY=YOUR_GEMINI_API_KEY

Run the application

streamlit run app.py

---

📱 How to Use

1. Upload a CSV dataset.

2. Select the target column.

3. The application automatically:

   • Cleans the data
   • Encodes categorical features
   • Detects the ML problem type
   • Trains multiple models
   • Compares performance

4. Generate AI-powered summaries.

5. View improvement recommendations.

6. Download the model results.

---

🌟 Why This Project?

Unlike traditional AutoML applications, this project combines Machine Learning automation with Generative AI.

Instead of only displaying evaluation metrics, it also explains the results and provides actionable recommendations for improving model performance, making it valuable for students, data analysts, and machine learning practitioners.

---

🔮 Future Enhancements

• XGBoost and LightGBM support
• CatBoost integration
• Hyperparameter tuning
• SHAP explainability
• Feature importance visualization
• Cross-validation
• Model deployment
• PDF report generation
• One-click model download
• Multiple dataset support

---

👨‍💻 Author

Vikash Basfore

• Data Science | Machine Learning | Generative AI
• Python • Scikit-learn • Streamlit • Google Gemini

---

🙏 Acknowledgements

Special thanks to the developers and communities behind:

• Google Gemini
• Streamlit
• Scikit-learn
• Pandas
• NumPy
• Python Open Source Community

Their incredible tools made this project possible.

---

⭐ If you found this project helpful, don't forget to Star the repository!