Sign In

Communications of the ACM

ACM TechNews

Researchers Mine Millions of Metaphors Through Computer-Based Techniques

View as: Print Mobile App Share:

A digital humanities project that started at Stanford University is teaching computers how to analyze text and extract metaphors to build a searchable database, which will enable users to browse historic patterns of word uses.

"As a tool, it provides a really powerful way of thinking about a lot of literature at once," says University of Virginia English literature professor Brad Pasanek, who is working on the project with computer scientist D. Sculley. The project is making tangible what the German linguist Herald Weinrich called the "metaphoric field."

Digitized libraries have made a wide range of books available online, including obscure and rare books. Researchers can use data-mining techniques to search the millions of words in those books to study subtle changes in how the words have been used throughout history. "The nature of metaphor is such that it does not lend itself to easy detection by the usual sorts of pattern-matching algorithms," says Stanford computer scientist Matt Jockers, who created the digital database used for the project. Pasanek provided the computer with examples of metaphors and trained the software to recognize metaphors, using proximity searches between words likely to be metaphoric.

Sculley says a similar technique is used in spam-detection software. "Pasanek's database is the first metaphoric field that we can actually see and use," says Stanford professor Franco Moretti. "It provides empirical proof for a daring but never wholly solid concept."

From The San Jose Mercury News
View Full Article


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account