Sarcasm Detection of Amazon Alexa Sample Set

Abstract

Sentiment analysis using collection of positive, negative score of a word has been one of the most researched topics in Data Mining. This kind of analysis is more prominent based on the content available on social media like comments on Facebook, tweets on Twitter, and the count goes on. Sarcasm can be understood as irony but it is a text spoken in such a manner that evokes laughter and humor. It is a type of sentiment where people express their negative feelings using positive or intensified positive words in the text. While speaking, people often use heavy tonal stress and certain gestures clues like rolling of the eyes, hand movement, etc., to reveal sarcasm. In this paper, NLTK has been used which is a Python toolkit to harness the power of generating information from the huge text datasets available. Sampled data from Amazon Alexa has been collected which is further processed using SentiWordNet 3.0 and TextBlob to remove noise and irrelevant data. Thereafter, Gaussian naive Bayes algorithm along with TextBlob has been used to detect sarcasm in dataset. The performance of the proposed method is compared with naive Bayes, decision tree, and support vector machine. From the experimental results, effectiveness of the proposed method is observed.