Autism Prediction ML Web App

A machine learning web application for autism prediction using Logistic Regression, SVM, and XGBoost models. The project includes a Flask-based frontend where users can enter required details and get a prediction output through a simple web interface.

Project Overview

This project uses machine learning techniques to predict autism-related outcomes from structured data. It was developed as an academic minor project to demonstrate how trained ML models can be integrated with a web application.

The project includes model training, saved machine learning models, dataset files, and a Flask-based frontend for real-time prediction.

Features

• Autism prediction using machine learning
• Logistic Regression model
• Support Vector Machine model
• XGBoost model
• Flask-based web application
• User-friendly prediction form
• Saved trained models using Joblib
• Jupyter Notebook for model analysis and training
• Dataset files for academic reference
• Confusion matrix visualization included

Tech Stack

• Python
• Flask
• Pandas
• NumPy
• Scikit-learn
• XGBoost
• Joblib
• HTML
• CSS
• Jupyter Notebook

Project Structure

Autism Prediction.ipynb
train.csv

Frontened Autism Prediction/
│── app.py
│── requirements.txt
│── train.py
│── readme.md
│── toRun.txt
│── modules.txt
│── confusion_matrix.png
│── feature_columns.joblib
│── feature_scaler.joblib
│── logistic_regression_model.joblib
│── svm_model.joblib
│── xgboost_model.joblib
│
│── data/
│   │── train.csv
│   │── test.csv

How to Run the Project

Follow these steps carefully on Windows using VS Code.

Step 1: Clone the Repository

git clone https://github.com/tiwari369/Autism-Prediction-ML-Web-App.git

Step 2: Open the Project Folder

cd Autism-Prediction-ML-Web-App

Step 3: Go to the Frontend Folder

cd "Frontened Autism Prediction"

Step 4: Create a Virtual Environment

python -m venv .venv

If the above command does not work, try:

py -m venv .venv

Step 5: Activate the Virtual Environment

For Windows PowerShell:

.venv\Scripts\activate

After activation, the terminal should show something like:

(.venv)

Step 6: Upgrade pip

python -m pip install --upgrade pip

Step 7: Install Required Libraries

pip install -r requirements.txt

If any package error occurs, install the common required packages manually:

pip install flask pandas numpy scikit-learn xgboost joblib

Step 8: Run the Flask Application

python app.py

Step 9: Open the Web App in Browser

After running app.py, open this link in your browser:

http://127.0.0.1:5000/

Now fill the form and submit it to get the autism prediction result.

Model Files Used

The web app uses trained model and preprocessing files stored inside the frontend folder:

logistic_regression_model.joblib
svm_model.joblib
xgboost_model.joblib
feature_scaler.joblib
feature_columns.joblib

These files help the application load the trained models and make predictions without retraining every time.

Training File

The project also includes:

train.py

This file can be used to train or regenerate model files if required.

To run the training script:

python train.py

Dataset

Dataset files are included for academic learning and model testing purposes:

train.csv
data/train.csv
data/test.csv

Common Errors and Fixes

1. Flask not found

If you get:

ModuleNotFoundError: No module named 'flask'

Run:

pip install flask

2. XGBoost not found

If you get:

ModuleNotFoundError: No module named 'xgboost'

Run:

pip install xgboost

3. Joblib model file not found

Make sure you are running the app from inside this folder:

Frontened Autism Prediction

Correct command:

cd "Frontened Autism Prediction"
python app.py

4. Virtual environment activation issue

If PowerShell does not allow activation, run:

Set-ExecutionPolicy -Scope CurrentUser -ExecutionPolicy RemoteSigned

Then activate again:

.venv\Scripts\activate

Files Not Included Publicly

Private academic documents such as reports, presentation files, and tool/source documents are intentionally excluded from this public repository using .gitignore.

Excluded file types include:

*.docx
*.pptx
*.pdf

Use Case

This project is useful for:

• Academic minor project demonstration
• Machine learning classification practice
• Flask ML deployment learning
• Healthcare-related ML project understanding
• Model integration with a web frontend

Author

Riteek Raj Tiwari

Disclaimer

This project is created only for academic and learning purposes. It should not be used as a real medical diagnosis tool. Autism-related concerns should always be discussed with qualified medical or healthcare professionals.