TCSD: Term Co-occurrence Based Sarcasm Detection from Twitter Trends

摘要

The scope of social media platforms like twitter, enabling the target audience to surge their opinion about the vivid domains like products, entertainment, promotions of a concept, government related decisions, individual celebrities and many more. Sentiment analysis is the potential strategy of data science and data mining, which allows notifying the opinion of the target audience. The opinions projected in social media platforms often adapt sarcasm to portray the opinion with negative sentiment polarity. In such context, the computer aided methods of sentiment analysis prone to considerable level of false alarming. The contributions of contemporary research in sentiment analysis evincing that the detection methods of sarcasm in user opinions is a crucial task of the sentiment analysis. In the context of this argument, a novel machine learning based method is derived to identify the sarcasm in the user tweets. The simulation study shows the proposal importance, which measured by comparing the performance of the proposed model with another contemporary model.