Catriple
Catriple is a system that works to automatically extract triples about Wikipedia articles and its non-isa properties from Wikipedia categories.
The data extracted by our methods complements existing semantic web data which is the basis for advanced applications and realization of the Semantic Web. This is a fundamental step of our entire work. The extracted data can be used in several applications such as realizing semantic search for Wikipedia, enriching Wikipedia infobox data, refining Wikipedia category system, etc. In fact, we have already begun to implement these applications. For instance, an online demo of semantic search for Wikipedia based on the extracted data is provided here. (You can input keywords "2008 films" to have a try or see this screen_camera.avi). Although it is still under construction, it can be a potential mechanism to improve access to the large Wikipedia knowledge base. The data can also help tasks in other fields, e.g. question answering, information retrieval, information integration.