Friday, October 19, 2007

Impress Your Professor: Image control or not to control

Images are multiplying by the billions on the internet and it's folks like us who must figure out how it is, exactly, that we are to retrieve all of this information! Images, however, pose a unique challenge of context: cultural, visual, linguistic and otherwise. Dr. Choi, illustrates the complexity of the matter with this picture:

To control or not to control the vocabulary...that is the question.

This summer, the Wall Street Journal featured an article on the subject: Computer Scientists Pull a Tom Sawyer To Finish Grunt Work (June 27, 2007; Page B1). The gist was that computer scientists have come up with a new way to tackle the massive task of classifying images in large databases by making a game of it.

In this game, two random users are shown one image and are asked to type in words to describe it. When their words match they "win" and the word is chosen as a descriptor on the assumption that there is at least a certain level of congruency. This is called free indexing and is based on the Web 2.0 (see Flickr) concept of collaborative tagging.

This method leads to a lowest common denominator situation and the classification will tend to remain relatively basic.

(For instance, it wouldn't really work for images of, say, the medulla oblangata, because who knows what the heck THAT looks like, and we'd all just end up tagging a picture of it as 'brain'. or perhaps 'mushy sponge'.)

Labeling images is not a process so easily resolved by games but the idea is extremely interesting and of course not without its detractors.

In the October/November edition of the ASIS&T Bulletin the debate over how to classify images continues as Elaine Ménard asks Image Indexing: How Can I Find a Nice Pair of Italian Shoes? She makes a case for traditional controlled vocabulary indexing and argues that there is no evidence that there is a "reason to consider collaborative tagging as a replacement solution to traditional indexing with controlled vocabulary". She argues that indexing is best left up to the experts....

and the debate rages on!

On November 7th, in the lovely May Gallery at Mullen Library, come see Dr. Youngok Choi present "Searching for Books and Images in OPAC: Effects of LCSH, TOC and Subject Domains" a study done by Dr. Youngok Choi, Dr. Ingrid Hsieh-Yee and Dr. Bill Kules of SLIS.
Admission if free for students with registration!

No comments: