Exposing Latent Information in Folksonomies for Reasoning

In this project, we propose to develop an approach for identifying and exposing the latent semantics within a folksonomy, which will in turn enable a new class of data integration applications. In previous work, we developed software enabling non-programmers to create web feeds, and a portal system for displaying that data in an integrated view. The new application we are proposing to develop in this project will enable domain-experts to automatically integrate webfeeds into the portal without any programming being required. To achieve this, we will be investigating the use of statistical relational learning to learn classification rules based on folksonomy metadata. The domain expert can then train the system to perform the integration task autonomously.