Published October 20, 2025 | Version v1

AI-Based Disease Prediction System Using Machine Learning with Conversational Assistance and AI-Driven Recommendations

  • 1. BRACT'S Viswakarma Institute of Information Technology
  • 2. BRACT'S Vishwakarma Institute of Information Technology
  • 3. BRACT'S Vishwakarma Institute of Information Technolgy
  • 4. Vishwakarma Institute of Technology

Description

This paper presents a web-based medical diagnosis
assistant that predicts probable diseases from user-reported
symptoms using supervised machine learning. The model is
trained on a large dataset containing approximately 246000 rows
for diseases mapped across 377 symptoms. Classical machine
learning models including Na¨ıve Bayes and Random Forest
were implemented and compared. The Na¨ıve Bayes classifier
achieved the best overall accuracy in our experiments. The
final deployed system integrates an interactive web interface,
automated PDF report generation, a conversational chatbot
for natural-language symptom input, and Google Gemini-based
recommendation generation for remedies and precautions. We
discuss data preparation, model selection and tuning, evaluation
metrics, system deployment, ethical considerations, explainability
via SHAP/LIME, and future work. The system is intended as
a preliminary screening tool to improve health awareness and
help users make informed decisions before consulting healthcare
professionals.
Index Terms—Disease Prediction, Na¨ıve Bayes, Random For-
est, Machine Learning, Chatbot, Gemini API, Explainable AI,
Healthcare Informatics

Files

AI_Medical_Diagnosis (1).pdf

Files (630.1 kB)

Name Size
md5:0a3d9de4cb88203b569d8850ef2a7317
630.1 kB Preview Download

Additional details

Software

Programming language
Python , JavaScript , HTML , CSS
Development Status
Active