A Comparative approach for Sentiment Analysis from Summarized User Health posts

Abstract

Now a day there are number of online health communities provide a huge amount medical information that are posted by patients. The health post contains patients view, opinion or experience on particular drug. In this work we proposed to collect real time posts from websites like askapatient.com, healthboards.com etc. We proposed to summarize user posts using simplified Lesk algorithm. While summarizing the importance of each sentence is calculated using online dictionary called WordNet. Depending upon the percentage of summarization given by the user, the top ‘m’ sentences will be selected as summary. We proposed to classify users based on the sentiment expressed in their posts. To classify users we proposed to use Machine Learning Algorithms such as Naïve Bayes classifier and Support Vector Machine(SVM) and comparing both the classifiers.