Computation Cracks 'Semantic Barriers' Between Databases

Differences in vocabulary between fields make it difficult or impossible to search for a familiar concept in collections of data from an unfamiliar field, which is why reference librarians are invaluable. But on the Internet, where databases are proliferating and librarians are nowhere to be seen, users need another way to cross these “semantic barriers.” Now, by using a supercomputer to comb through millions of abstracts in different fields, searching for patterns of association between key terms, information scientists laid the groundwork for a system that can suggest keywords for finding known concepts in unfamiliar databases. Computers/Math