Published Paper Details:

Unstructured data remains to be a challenge in every data intensive application domains like business, research and technology driven companies and they are in the form of tweets, news, emails, reviews etc. Using text analytics we can extract meanings, patterns, and structure hidden inside unstructured text data. The term �text analytics� is an integrated framework by using techniques from data mining, machine learning, natural language processing (NLP). Text analytics uses techniques and methods that are used to get insights from unstructured data. These techniques can be broadly classified as topic extraction or modelling, cluster analysis which is a part of exploratory data analytics and sentiment analysis also called text classification which is a part of predictive text mining which is also called machine learning. To apply any kind of text mining technique like clustering or classification on text it has to be first pre processed and need to be converted to a bag of words model so that we can extract features form the pre-processed data where we convert documents into vectorized format using word frequency count which has values in binary form so the machine(computer) can understand. SMART URT extends for Smart Unified extraction of sentiments from reviews and topics from documents which is a natural language processing unified framework in which I built a topic extraction model using text cluster analytics and also built a sentiment analyser which predicts if a movie review is positive negative or neutral.