Google Developing 'PageRank' for Image Search
Last week at the International World Wide Web Conference in Beijing, Google researchers presented VisualRank, an algorithm that blends image-recognition software with techniques for ranking similar images, reports The New York Times.
VisualRank is positioned as PageRank for images. Google's PageRank helps determine a site's value, based on content and scaled from 0-10. The higher the PageRank, the higher the site appears in organic search listings for related keywords.
At an ad:tech session on search engine optimization in San Francisco two weeks ago, Google's Aaron D'Souza stressed the importance of incorporating alternative content to Google search results — including images and video, which he called "psychically rewarding" to users.
Images and videos that appear on a search engine are typically generated based on tags or surrounding text (a video title, for example). But there is currently no effective way to filter duplicates, or quantify the "value" of one versus another in a category.
To develop VisualRank, Google researchers focused on the 2000 most popular product queries on Google, including iPod, Xbox and Zune. They later determined the top 10 images from its ranking system, gleaned in part from Google Image Search results.
Employees then created a scoring system for image "relevance." The system returned 83 percent fewer irrelevant images.
Return to marketing news headlines
View Marketing News Archive