A Novel Ml Based Approach for Clinical Information Extraction Based on Lexical Features and Candidate Pairs

EMRs, or electronic medical records, are the main resources that make it possible to reuse clinical data. Since EMR data is free text, information must be extracted using natural language processing techniques. The main element that makes a difference in utilizing the valuable data found in the Information Extraction Task is its semantically focused functionality. Systems that use natural language processing (NLP) have the ability to carry out IE tasks. Such an NLP system would, however, produce inconsistent outcomes and always leave room for development. There may be room for improvement in the Clinical Information Extraction Process because the semantics of information extraction in EMRs will be dependent on temporal expression. Therefore, in this work, we offer an improved descriptive methodology that makes use of lexical-based clinical text extraction.