Disease Predicting Web Application Using Machine Learning
Sreeraj P1, Ms. K.V. Indulekha 2
1II MCA, Department of Computer Applications, Nehru Institute of Information Technology and Management, Coimbatore, Tamilnadu, India
2Assistant Professor, Department of Computer Applications, Nehru Institute of Information Technology and Management, Coimbatore, Tamilnadu, India
Abstract - In recent years, the integration of machine learning into healthcare systems has gained significant momentum, particularly in areas focusing on early diagnosis and predictive analytics. This research presents the design, development, and evaluation of a multi-disease prediction web application that leverages binary classification models to identify potential illnesses based on user-provided symptoms. The system is built upon a modular machine learning architecture, where each disease — including AIDS, Allergy, Dengue, Diabetes, Heart Attack, Jaundice, Malaria, Pneumonia, Tuberculosis, and Typhoid — is addressed through an independently trained classifier using labeled datasets. This approach enhances model precision by isolating disease-specific symptom patterns and mitigating multi-class confusion.
The application provides an intuitive user interface, developed using Gradio, which supports both individual and batch predictions via manual symptom input or CSV file uploads. It returns predictions with associated confidence scores and also suggests relevant diagnostic tests to aid further clinical validation. User authentication mechanisms have been incorporated to ensure data privacy and secure access. The backend is implemented using Python-based frameworks such as Flask and Scikit-learn, and the entire application is deployed on Hugging Face Spaces for cloud accessibility and scalability.
This project shows how machine learning can help with early disease detection, especially in areas with limited medical access. It's easy to use, expandable for future features, and supports both individuals and healthcare workers in making quick, informed health decisions.
Key Words: Machine Learning, System-based Diagnosis, Healthcare AI, Gradio Interface, Scikit-Learn
