Entity Identification in Deep Web

The Web has been rapidly deepened with the prevalence of databases online. From different Web sources, the records usually use different representations to refer to the same real world entity. Therefore, there is a high demand for identifying these entities from multiple Web databases in many Web application, e.g., comparison shopping. In this paper, we propose an effective entity identification approach which is based on a similarity function. Moreover, we develop query-based dynamic weight techniques in our approach. Experimental results show that our approach can effectively discover the records representing the same entity in the real world.