Sarcasm is a special form of irony by which the person conveys implicit information, usually the opposite of what is said, within the message he transmits. Sarcasm is largely used in social networks and microblogging websites, where people mock or criticize in a way that makes it difficult even for humans to tell if what is said is what is meant. Recognizing sarcastic statements can be very useful when it comes to improving automatic sentiment analysis of data collected from social networks. It helps also enhance the efficiency of after-sales services or consumer assistance through understanding the intentions and real opinions of consumers when browsing their feedbacks or complaints. In this paper we propose a method to detect sarcasm in Twitter that makes use of the different components of the tweet. We propose four sets of features that cover different types of sarcasm we defined, and that will be used to classify tweets into sarcastic and non-sarcastic. We evaluate the performances of our approach. We study the importance of each of the proposed sets of features and evaluate its added value to the classification.