SmartScan: a comparative analysis of Roberta and Bert for disease prediction from patient symptom profiles

Machine Learning (ML) and AI has evolved a lot and influenced different domains including medicine, language treatment, text recognition and many other domains. The paper explores the use of ML alternatives aiming to diagnose diseases, automatic text analysis, language processing and data extraction adopted by the website. Investigate the performance of models such as Nave Bayes, Support Vector Machine (SVM), K-Near Nachbarn (KNN), Random Forest, Bert, Roberta, XLMRoberta and TROCR. This research gives idea about using ML for lung disease prediction better text processing via part of Indic language (POS) trustable optical property prediction (OCR) and how web scraping of ecommerce information is best done via AI controlled knowledge transactional efficiency. Applying machine learning in these domains enables breakthrough in medical diagnosis, language understanding, text recognition and in real-world market based analysis.