Sign In

Communications of the ACM

[email protected]

Collective Intelligence: a Fad or Real Research?


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Ed Chi

Ed H. Chi - Google Research Scientist

One enduring core value in Human-Computer Interaction (HCI) research has been the development of technologies that augment human intelligence. This mission originates with V. Bush, Licklider, and Engelbart, who inspired many researchers such as Stuart Card and Alan Kay at PARC in the development of the personal computer and the graphical user interface.  The aim of augmented human cognition has remained a core value for Human-Computer Interaction research. 

A natural extension of this idea in the Social Web and Web2.0 world is the development of technologies that augment social intelligence.  Here I'm referring to the development of Social Web technologies that enable people to better foraging, share, tag, and make sense of complex topics.  Scenarios of this social intelligence include the use of forum to discuss  treatment options for cancer patients, political junkies perusing the blogsphere to track opinions, etc.

Social Foraging
The Social Web is enabling people to find and search for information more efficiently than before.  By using voting or other collective averaging methods, websites are able to deliver information with better signal-to-noise ratios. Digg.com and StumbleUpon.com, for example, both use voting schemes to identify the most popular and interesting news items.  Each item is submitted by one of the users.  Google’s original PageRank algorithm sifts through web pages, looking for links that vote for other pages.  The writers, essentially vouching for the contents of the linked pages, were participating in a collective averaging system.

Social Sharing and Tagging
Users wanting to share photos and bookmarks with others started tagging document objects with keywords. Users upload and share various media content with each other through these websites.  In many social tagging systems, sometimes the user motivation was more personal, such as simply wanting to remember interesting photos or websites for later consumption or re-finding. Flickr.com is a good example here, and I've used it to find interesting tourist sites to visit for my Cambodia trip.

The effect is the same.  The tags form a new and shared vocabulary that can be used to find and re-find interesting and popular content.  Data-mining algorithms are developed to cluster photos, videos, and documents using these tags. The result is a collaboratively formed information structure that can be used to navigate, search, and browse through the myriad of contents.

Social Sensemaking
Perhaps what’s most surprising was the willingness of web users to participate in open-source style creation of new complex contents.  As a giant experiment, the use of Wikipedia as a depository of knowledge about all kinds of topics was simply stunning.  Users writing articles about diverse subjects from abortion rights and the invention of radio to the cause of windburn, wiki-based system is used to make sense complex materials and resolve conflicts.

In the coming months, I will blog for CACM here to outline the latest research happening in this space, and I welcome your input.
 


Comments


Peter Brusilovsky

Ed, thanks for a great start! Looking forward to reading your future posts. But how about social navigation and social search? Is it something hidden in one of your three categories or you do not really consider these two as forms of Collective Intelligence?


Ed Chi

Peter:

I lumped social navigation/browsing and social search under social foraging. In fact, I recently wrote about how "information seeking can be social" in a recent publication:
Chi, Ed H. Information Seeking Can Be Social, Computer, vol. 42, no. 3, pp. 42-46, Mar. 2009, doi:10.1109/MC.2009.87

So absolutely, I hope to write some about that topic here as well.


Displaying all 2 comments