Conceptual Level Semantic Sentiment Analysis Using Twitter Data

Sentiment analysis (SA) AKA opinion mining is one of the frequently used Natural Language Processing (NLP) application to identify human opinions from their social media posts or reviews. Its ultimate goal is to identify the hidden emotions from their posts. The mission of SA is the classification of the opinion from the text or message. The main task of SA is to recognize and gather human opinions on various objects, topics and events. In this paper, semantic concepts are added as extra feature to the trained data set in sentiment analysis for more accurate classification of the result. We applied conceptual semantic sentiment analysis to train the three different twitter data sets. Using this notion, we can forecast the sentiment of any tweet which does not have sentiment keywords. The main emphasis is to increase the accuracy of twitter sentiment analysis by integrating the semantic analysis of words into sentiment detection.