AYURSOFTAPP- Software That Suggests Drugs and Formulations for Disease/Pharmacological Property Based on the Ayurvedic Classical Books/Repositories

AYURSOFTAPP- Software That Suggests Drugs and Formulations for Disease/Pharmacological Property Based on the Ayurvedic Classical Books/Repositories.

1Dr.Prof. Sukruth Gowda M A, 2 Ms. Harika Y, 3 Mr. Shyam R, ⁴Mr. Samuel R, ⁵Mr. Praveen Kumar G S

1Assistant Professor, 2Student, 3Student, 4Student, 5Student

1School of CSE & IS

1Presidency University, Bengaluru, India.

Abstract

 This project presents a modern, intelligent approach to Ayurvedic diagnosis and treatment recommendation by integrating Machine Learning (ML) techniques into traditional medicine. The goal is to bridge the gap between ancient Ayurvedic wisdom and contemporary technological advancements, making holistic healthcare more accessible and personalized. Built as a Flask-based web application, this system allows users to input general symptoms in natural language. Using this input, the system predicts relevant Ayurvedic symptoms and recommends suitable herbal treatments. Ayurveda, a time-tested system of natural healing, relies on the balance of the three doshas—Vata, Pitta, and Kapha—to diagnose and treat illnesses. However, understanding

Ayurvedic symptoms and remedies often requires years of training, making it less accessible to the general population. With the growing interest in alternative and traditional medicines, especially those grounded in natural treatments and preventative care, there is a need for tools that make Ayurvedic knowledge more user-friendly and data-driven. This project aims to fulfil that need by using Machine Learning models to assist in symptom classification and treatment suggestions.

The core of the system utilizes Natural Language Processing (NLP) to process free-form symptom descriptions provided by users. These inputs are first cleaned and transformed into structured formats using techniques such as tokenization, stop-word removal, and vectorization. The processed text is then fed into a Naive iv Bayes classifier, which has been trained on a curated dataset of Ayurvedic symptoms and corresponding treatments. The model predicts the most probable Ayurvedic symptom category and retrieves herbal medicine suggestions based on traditional Ayurvedic texts and clinical data.