Abstract: Search engines subvert how people think about the data in WWW. Since the data is massive, one of the most important goals today is to provide a service which people could search the useful information they want “efficiently”. Nevertheless, in some scenarios, it’s unrealistic to expect that a user to devise a set of keywords for getting the information from the web. For example, it’s impossible To figure out keywords you didn’t know yet. If you want to search good basketball players in the NBA, it’s difficult to define keywords for search without knowing the name of players or knowledge of basketball games. We tend to overcome these drawbacks by providing the users with abilities to search by examples. The goal of this project is to provide an approach for web users to search similar objects from the webs, instead of using keywords to define a concept they want (traditional search). We will present an efficient algorithm to identify entities under the assumption that tag information is given. A live demo will also be demonstrated during the presentation. Biosketch: Chu-Cheng Hsieh is currently a research intern at core research team in Symantec research lab, and he is also a graduate student researcher (enrolling in PhD program) in Computer Science department of University of California, Los Angeles (UCLA), supervised by Professor John Cho. Before joining UCLA, he was enrolled in PhD program in Computer Science department of National Chia Tung University, Taiwan, between 2005~2007. Also, he got a M.S degree and a B.A. degree in Electronic Engineering department from National Taiwan University of Science and Technology in 2005 and 2001 respectively. With respect to his working experience, before entering UCLA, he was an researcher in Telecommunication Laboratories of Chunghwa Telecom Co., Ltd, Taiwan, and served as a lecturer in mutilple universities in Taiwan. His current research interests include data mining, information retrieval, security, and social networks.