Sarcasm detection in natural language processing

The striking property of satire is that it makes it difficult to bridge and bridge the gap between its literal and intended meaning. Identifying sarcastic behavior in the field of online social networks such as Facebook, Twitter, Instagram, surveys, etc. has turned into a fundamental task as they affect social and personal relationships. SARCASM detection is an important processing problem in natural language processing (NLP), which is needed for better understanding to serve as an interface for mutual communication between machines and humans. To understand this is to underline the basic problem behind it - being able to detect the contradiction. For this, a need arose to define contextual understanding and emotion. To accomplish this we need to do two things - gather a stack of target words that display sentiment shifts (sarcastic words) based on context; And with an objective word given an expression, how to naturally identify whether the objective word is used in an exact or sarcastic sense. Collecting information is done by the use of an information retrieval system, for example tweepee. For the latter, some distributed semantic methods are used to convert data into useful information and are then demonstrated using multiple classifier results that is, satire is identified.