Framework for enhancing the Emotions of EMR using Ontology in Sentiment Analysis
By analyzing a data using NPL, electronic health records (EMRs) could offer many insights, which have not been exploited yet .Through Sentiment Analysis we were able to present the issue in a way that the physicians could identify with and solve. If the stored records are analyzed using a Natural Language Processing methodology (NLP) it will be very helpful in automating the process of collecting ,analyzing the data. This work aim to predict the International Classification of Diseases, Revision 10(ICD-10) code(s) – or it’s (their) derivatives – from the raw text records. Through we can easily diagnose the disease based upon the patient’s foretelling symptoms instead of going each and every long written data through nurses or its previous medical history. In this paper we represent the pipeline approach on information extraction, sentiment analysis, creating ontology for unsupervised learning and summarization technologies. Sentiment Analysis is performed through recursive neural deep learning and lexicon analysis. In this paper Ontology has the major concern to provide better prediction of related diseases and helps in more proficient summarization. The feasibility of the approach is evaluated through linguistic analysis and user studies. In the presented work we also summarize the effectiveness of the automated EMR against the traditional EMR.