An Enhanced Review Classifier for Analyzing User Sentiment

Abstract

Now days, fast developing of web applications, sentiment analysis would be big opportunity to measure analysis of user’s reviews from web information. Sentiment Classification mainly used for analyzing certain event’s or product based on the positive or negative opinions.  Dynamic opinion mining will be huge benefit for both normal people and product buyer. Still now, it is a complicated work and big issue. To avoid this, we proposed POS classification based Sentiment analysis. To increase querying time complexity during run time meta data analysis and requires having a remote to initiate content POS requests. Here, we propose to replace the HowNet api with an open-source entropy based proposed POS algorithm that comes with an max-net that will generate similar Pos’s quickly and efficiently.