As the number of information sources accessible on the Web grows, the
ability to execute queries spanning multiple heterogeneous sources is
becoming increasingly important.  One of the key challenges in providing
such a capability is to discover the semantic correspondences between schema
and data elements across the information sources. Most previous solutions to
this problem rely in some fashion upon identifying syntactic similarity or
identifying common domains of values in schema instances in order to
discover the correspondence between them.  While these approaches are
valuable in many cases, they are not infallible, and there exist instances
of the schema matching problem for which they do not even apply. Such
problem instances typically arise when the column names in the schemas and
the data in the columns are "opaque" or very difficult to interpret.
Addressing this problem, we introduced an automatic schema matching
algorithm, "un-interpreted matching," that does not rely upon the
interpretation of data. This talk will attempt to introduce the
un-interpreted matching method and present some of its recent developments.