OCR Text Recognition and Recommendation Using Machine Learning

The proposed system creates Ayurvedic and pharmaceutical alternatives using sophisticated text extraction and analysis methods. Text is extracted from PDFs or images using OCR, and fuzzy matching is applied to identify ingredient names correctly by mapping them to their actual forms. More advanced techniques like KNN and decision trees are used to propose different methods for a drug's composition and action. NLP boosts data extraction and interpretation by connecting diseases with relevant drugs, making recommendations more accurate. The system is based on a light-weight Flask backend for processing user requests, querying a structured database, and producing context-sensitive recommendations. Besides providing Ayurvedic medicine-based options, the platform broadens patient options, enhancing healthcare solution accessibility and affordability. Several pre-processing methods, such as noise reduction and image alignment, are used to maximize OCR accuracy. When faced with low OCR outputs or user errors, fuzzy matching again enhances text recognition. With its ability to offer dependable and informed healthcare options, this site has the capacity to transform the accessibility of healthcare by providing pharmaceutical as well as traditional medicine alternatives.