This thesis is all about how we can use a special tool called Named Entity Recognition (NER) to help us extract important information from Electronic Health Records (EHRs), like details about diseases and medications. By using a smart computer model called a NER Tagger, which is really good at picking out different types of information in text, we can make sense of all the medical jargon in these records. We used a powerful method called Conditional Random Field (CRF) to train this model on a bunch of medical text and then tested it to make sure it works well. The cool thing is that if we can do this accurately, it could make organizing medical records much easier and help us understand how diseases and treatments are connected. Overall, this research could make a big difference in how we handle and learn from medical data.