Prediction on Sarcasm Sentiment Detection of Twitter Data

EasyChair Preprint 2870

Abstract

This Analysis mainly predicts the sarcasm sentiment detection of  twitter. As detecting sarcasm in sentimental analysis is one of the most challenging task. Opinion mining is adopted for this study which covers various linguistic phenomenon such as positive, negative, sarcastic, ironic etc. Where it is possible to analyse the sentiments for text data from a very popular micro blogging website Twitter. Sarcasm is a type of dialect where ordinarily, the speaker expressly states the opposite of what is actually meant. Instilled with purposeful equivocalness and nuance, detection sarcasm is a difficult errand, even for human beings. This is due to the absence of vocal prompts and outward appearances and is generally lost in the content. For humans it’s very difficult to identify the vocal inflection. Sarcasm detection without vocal cues is very complicated task in hand. The existing state of art solutions in sentimental analysis and sarcasm scope  detection has been investigated. Moreover, a corpus of social media data with linguistic negation has been developed and an enhanced framework for sarcasm detection has been developed and assessed .And the results are not accurate by this means making it a machine learning algorithm to detect sarcasm sentiment. Hence, proved that by including a combined approach of hyperbole, emoticons, lexical analysis, contrast can achieve better accuracy when compared to usage of only linguistic features. However, exactness and strength of results are frequently influenced by false sentiments that are of sarcastic in nature and this is regularly left unnoticed. Designed a machine learning algorithm for sarcasm detection in content by utilizing the existing work and add improvisations on it. By breaking down the qualities and shortcomings of the existing models, it is to develop a new  model that will accomplish better results.

Keyphrases: Machine Learning Algorithm, Sarcasm Detection, Sentimental Analysis, features

@booklet{EasyChair:2870,
  author    = {Krishna Raj and Sireesha Polamuri and Sayyaparaju Sai Durga Vijaya Preethi and Sneha Penmetsa and Susmitha Pilli},
  title     = {Prediction on Sarcasm Sentiment Detection of Twitter Data},
  howpublished = {EasyChair Preprint 2870},
  year      = {EasyChair, 2020}}