Autism Prediction

This project aims to predict Autism Spectrum Disorder (ASD) using machine learning techniques. It involves data exploration, model training, and provides a web-based interface for making predictions.

Web interface for the Autism Prediction project

Key Technologies/Skills

  • Python & Jupyter Notebook
  • Scikit-learn, Pandas, Numpy
  • Matplotlib & Seaborn
  • Flask (Web Framework)
  • HTML & Pickle
  • Data Exploration & Visualization

What I Accomplished

  • Managed an end-to-end ML workflow from data exploration to deployment.
  • Cleaned, transformed, and prepared data for modeling using Pandas.
  • Trained and evaluated a classification model with Scikit-learn.
  • Saved and loaded the trained model using pickle for persistence.
  • Developed a basic web application with Flask to serve predictions.

Notable Features

analytics

Interactive Data Exploration

Jupyter notebooks provide a platform for detailed data analysis.

model_training

Machine Learning Model

A trained and serialized model for autism prediction.

web

Web Prediction Interface

A user-friendly web application for making live predictions.

Future Improvements

  • Implement robust error handling and input validation in the web app.
  • Explore more advanced models or ensemble methods for higher accuracy.
  • Containerize the application with Docker for scalability.
  • Automate testing and deployment with a CI/CD pipeline.

Project Highlights

  • Successful implementation of an end-to-end ML project.
  • Demonstrates skills in data science and web development.
  • Provides a practical application of ML for a health domain.
View on GitHub ← Back to Portfolio