Trait Specific Sentiment Analysis Using Natural Language Processing (NLP).
We present a novel and generic approach to go about online shopping. Through our project, user can not only buy products across various websites, but also be able to know which product suits him better. Using scraping and data preprocessing to get reviews, we find popular traits in the product categories using Natural Language Processing Tools. We then ask user’s preferences about these traits. Further, we use our own algorithms to design a system which analyzes those reviews to match them with user’s priority against those specific popular traits. The system learns the set of significant traits by itself which allows us to extend it to any field. The data requirement is minimal as this is a one-time parsing of reviews.
In this particular project, we show how this could be applied in case of mobile phones, but this system and algorithm being generic, is domain independent and will change according to products chosen.