Sarcasm detection is a narrow research field in natural language processing (NLP) and the specific case of sentiment analysis. Sarcasm detection identifies irony and incongruous containing utterances in the sentimental text from the user. The significant goal of sarcasm detection is to detail understanding of user-s opinions and sentiment. A widely applied automatic detection model for sarcasm in NLP is the deep learning model.
  Deep learning can automatically detect discriminatory features in sarcastic sentences. Sarcasm detection is a difficult task for the reason that largely dependent on contextual knowledge and accent in the sentences. The rule-based model is incorporated with deep learning to overcome the difficulty, as it contains a learning classifier system with a set of rules to cover the contextual knowledge. Integrating deep learning with a rule-based model for sarcasm detection efficiently extract and gather meaningful features from the ironic utterances.