Swarm intelligence is the theory that there are a set of rules that define behavior and that those rules can be distilled to an algorithm. The algorithm can then be used to find good solutions quickly. According to computational scientist Xiaohul Cui of Oak Ridge National Laboratory, swarm intelligence can yield fast, although frequently approximate, solutions. But an approximate solution is often better than the best possible solution. If you were trying to evacuate a city, for instance, faster is definitely better than perfect. Perfect may not even work.
Working with Jessie St. Charles of University of Tennessee at Chattanooga, Cui is using swarm intelligence to assist both the United States Navy and Air Force in document organization. Swarm intelligence allows them to "cluster" information for easier retrieval by linking according to relevance to specifics.
According to St Charles and Cui, there are specific rules that define a swarm or flock, and because those rules can be quantified, they can be linked and applied.
The first rule is the rule of separation. Think of a flock of birds flying together. They fly in close formation but never so close that they collide. Separation not only draws them close but also introduces a repulsion force to insure that. The second rule is cohesion. Back to our flock, the birds in that formation don't want to get too far from their neighbors. The third rule is alignment. This allows a bird to gauge where everyone is going and align themselves to the group. The fourth rule, one that Cui added a few years ago, is species or type recognition. Our flock of birds are all mallards. A Canada goose would not join that flock.
In terms of the database, documents are stripped of non-meaningful words and word endings then analyzed for frequency of remaining terms. This results in an ID that can be used to assess relatedness. The end result of this is a faster aggregator with more relevant results.
While St Charles and Cui's work focuses on search, the hive mind or swarm intelligence theory has been with us for some time. Charles Leadbeater points out in "We-Think" that there is an exponential growth in collective thinking and imagining that has been fostered by the information revolution.
In We-Think, Mr. Leadbeater proved his theory by submitting an early draft of his book online and used the responses to his work to draft the final manuscript. He claims that this process is a template for the future — rather than a "top down" in which the writer speaks to the reader. The process is more collaborative, resulting in a final work that is more reflective of the audience's needs.
In the past three months, we have seen another application of how the hive mind can not only speak but take action. The group "Anonymous" has proven that by using the collective input of all who are willing to speak up, they have managed to mobilize a world wide effort focused on demanding a critical review of the Scientology and asking government leaders to review the tax exempt status that they currently hold. Within a very short time frame, Anonymous was able to organize people world wide for peaceful protests and have managed to hold those protests on a monthly basis. Scientology, meanwhile has tried unsuccessfully to find the leaders so that they could employ the usual tactics of litigation and harassment. Unfortunately for them, a collective mind has no leadership.
It is undeniable that there is power in the collective mind. The question that we need to consider is how we can use that power and use it responsibly.
What business application do you see in this? Is there a value in tossing a question to a wider audience? Isn't TechRepublic a type of collective mind, after all?
We-Think better than I-Think (The Telegraph)