Sarcasm Detection Model Building with Vector Visualisation

Abstract: Our work is building a sarcasm detection model which detects the sarcasm in a sentence that may be in news headlines or statements made on social media. There has been an exponential growth of social media in recent years. An immense amount of data is been put into the public domain through social media. To start this work first we have to know what Sarcasm is. When someone says or writes anything that is entirely unrelated to what they actually intended, they are expressing their feelings through sarcasm. Sarcasm is extremely difficult to spot because of how anonymous it is. The accuracy of the sentimental analysis can be increased by performing a perfect analysis and interpretation of sarcastic language. Understanding a person's attitude and opinions is the goal of the sentimental analysis. We attempted to describe the overall architecture of sarcasm, technique, and vector visualization of words using TensorFlow in this work.

Keywords: Sarcasm, Sentimental analysis, TensorFlow, Vector visualization, Social media.