MINING SERENDIPITY FROM DRUG REVIEWS USING SUPPORT VECTOR MACHINES

Serendipity is all about the Positivity or Goodness. Sentiment analysis is that the process of determining whether an editorial is positive, negative or neutral. Analysing opinions allows data analysts on large companies to guage vox populi, conduct nuanced research, monitor brand and merchandise reputation and understand customer experience. The proposed model is trained and evaluated with a UCI ML dataset that describes drug use. If the patient's drug use effect might be computationally identified, it could help generate and validate drug repositioning hypotheses; Measures are often taken for that drug, both in production and in use. We investigate supervised machine learning models to extract positivity in drug use. Furthermore, the proposed model is compared with Naive Bayes. Finally, a machine learning model using Natural Language Processing techniques on drug usage review dataset is implemented to find the sentiment of drugs and results were presented.