Facebooktwittergoogle_plusredditlinkedintumblr

For a long time, the conventional wisdom about tagging images was to use clear, precise words that interpret the picture exactly. For example, with the ubiquitous cat photo, you would use (cat, cats, kitten, cat lover, tabby). Pretty straightforward.

However, researchers propose that personalized tags can produce better search queries and recommendations. For example, a user might give a cat photo the tags (monty, pet, buddy, cat, chum). Clearly, the terms have personal meaning and can speak to what the user might really want to see in advertising and search results.

Researchers say the order in which the tags are ranked matter too. Currently, most search engines offer no direct way to learn and personalize a user’s preferences. Read more about how to change that in the April—June 2017 issue of IEEE MultiMedia.