Clustering uncertain data is an important task. Now-a-days for collecting information particularly reviews for a product is becoming important through web. As the number of customer reviews is increasing day by day, it is important to develop an efficient and effective technique to extract product features mentioned in customer reviews. In this work reviews in an unstructured text format is considered and online user review mining is done to generate feature-based result so that it helps a user while making purchase online. Pre processing and feature extraction is done on the reviews and further similarity between uncertain objects based on probability distribution is calculated.