Medicine Recommendation
Smart Health with AI
The problem Medicine Recommendation solves
This project is a medical recommendation system that helps users understand their health better by predicting possible diseases based on the symptoms they provide. The user can either type or speak their symptoms, and the system processes this input using a trained machine learning model. Once a prediction is made, the system not only shows the most likely disease but also provides useful information such as a description of the condition, precautions to follow, suggested medications, suitable diets, and even simple workouts. While building this project, I faced challenges like handling inconsistent symptom inputs, finding proper datasets, and connecting the backend with the frontend, but I resolved them through data normalization, creating structured CSV files, and carefully linking Flask with HTML, CSS, and JavaScript. This project can be improved further by expanding the dataset, using natural language processing for more flexible input, and integrating professional medical validation. The main benefit of this system is that it gives users quick awareness and preventive guidance about their health, showing how machine learning and web development can come together to create meaningful solutions in healthcare.
Challenges we ran into
One of the main challenges I faced was handling user input for symptoms. People often write symptoms differently, like “head ache” instead of “headache,” which didn’t match my dataset. I solved this by normalizing the text (lowercasing, removing spaces, and using underscores). Another difficulty was collecting reliable medical data, since most datasets were incomplete or inconsistent. To overcome this, I created structured CSV files for diseases, precautions, medications, diets, and workouts. I also struggled with model accuracy in the beginning, because some predictions were incorrect. I experimented with different algorithms like Random Forest and SVM, and finally chose SVM since it gave better results. Finally, connecting the frontend and backend smoothly was tricky at first, but Flask routes and careful debugging helped me integrate everything successfully.
Technologies used