Online reviews are popular among consumers because it allows them to rate businesses or products based on their personal experiences. Positive reviews can help generate more business as can negative reviews cause clientele loss. Unfortunately, this has been abused due to alternative motives such as the gain of profit and fame.  Products, hotels, and restaurants are main targets for online review deception. This paper studies fake review detection techniques to discover deceptive reviews in social media.  We compare three algorithms to classify texts and determine the most effective method with one being a Support Vector Machine (SVM), a supervised learning algorithm that analyzes and recognizes patterns in data. Our training data set consists of 400 truthful and deceptive reviews extracted from actual review websites. Choosing features such as bigrams, part of speech, and removal of stop words, we are able to build a SVM classifier with an 86% accuracy rate for fake review detection.