Medicine Recommendation System Using (NLP&GPT) AI

and accessible medical guidance. This study presents an AI-driven Medicine Recommendation and
Information System designed to support informed healthcare decisions, particularly for individuals
without immediate access to medical professionals. The recommendation engine uses a structured
dataset of medicines and related symptoms. Data preprocessing techniques—such as text
normalization, tokenization, stop word removal, and lemmatization—ensure consistent representation.
The processed data is converted into numerical form using TF-IDF and Count Vectorizer methods. A cosine similarity matrix then compares user-input symptoms with stored medicine profiles and ranks
medicines based on relevance. The system also includes a conversational AI module that allows users
to ask questions about dosage, usage, side effects, precautions, storage, and drug interactions. Through
intent recognition and a structured knowledge base, it provides clear and context-aware responses. With
a two-phase architecture consisting of offline processing and real-time deployment, the system ensures
efficiency, scalability, and improved digital healthcare support.