Correlation of feature score to overall sentiment score for identifying the promising features

Nowadays, most of the business intelligence focus on social media like facebook, twitter, blogs and online commercial websites like shopclues, pepperfry, flipkart, fabfurnish, testfreaks, amazon, greendust etc., to gather comments posted by the buyers in deciding about future demand,brand promotion, market segmentation and product penetration. In turn the buyers were also willing to post their comment about each of the products they buy through online. And these short reviews once refined and analyzed can help us to get a crystal clear opinion about the buyers' view which probably enhances the future buyers to make a buying decision based on spectacular features. This paper includes implementation on data acquisition, preprocessing, combinatory of lexicon and syntactic pattern mining approach (1) to find overall sentiment scores and correlate that score to that of feature score (2) to identify the most promising features of the product.