Searching the Web of Objects

摘要

We present a pragmatic approach to search the Web of Objects, that is, a Web where entities such as people or places are recognized and exploited. We outline a search architecture where information extraction and semantic technologies play key roles. This architecture has to cope with incompleteness as well as noise to expand the capabilities of current search engines. The main open problems for research are related with recognizing the entities in the query and ranking objects. We show some of these ideas through features or demos already available.