Ranking and interestingness

One of the notable things about Google Scholar is the citation-based ranking. There was some discussion recently about how Google was looking at new ways of ranking news articles, in part based on reputational characteristics of the source of the news. Now Flickr has announced its work on interestingness:

The other new feature is called interestingness and it’s huge! A long time in the making, interestingness is a ranking algorithm based on user behavior around the photos taking into account some obvious things like how many users add the photo to their favorites and some subtle things like the relationship between the person who uploaded the photo and the people who are commenting (plus a whole bunch of secret sauce). [FlickrBlog]

The characterization of interestingness, based on environment knowledge and user behavior, is becoming a major area of investigation and will become more so as we can find, manage and share more types of material on the web. As I suggest in the related entry below, it seems that we may see a growing differentiation of approach depending on the materials being handled, as the ranking technique developed for an interconnected world of web pages applies less to other types of material.
OCLC uses holdings counts to rank results. Circulation data provides an interesting opportunity to be mined for recommendation and ranking purposes.
Related entry:

One thought on “Ranking and interestingness”

  1. I suspect the “growing differentiation of approach” has more to do with the user than the materials. A student’s need for information might drive her directly to the classic, highly cited work to write a paper for a class assignment , but lead her to just browse the shelves to see what is “interesting” when she has a bit of spare time to let loose her curiosity.

Comments are closed.