Sentiment Analysis of Tweets Using Semantic Analysis

In today's world, there is endless stream of data present online and the automated analysis of such data holds a great promise in business analytics for providing a strong support in decision making. This paper looks at the very heart of the concept of sentiment analysis by classifying the tweets with the help of algorithms like Naïve Bayes, Maximum Entropy, and Negation. In this paper, we first preprocess the tweets to remove unnecessary content in tweet; we then extract the adjectives which forms the feature vector, which are also used to find synonyms used in further semantic calculations in aforementioned algorithms. Finally, we calculate Accuracy, Precision, and Recall to compare these algorithms.