Six degrees of separation in instant messaging
According to Nature in 'Six degrees of messaging,' computer scientists at Microsoft Research Redmond lab have logged a full month of instant messengers using -- logically -- Microsoft Messenger. 'The compressed dataset occupies 4.5 terabytes, composed from 1 billion conversations per day (150 gigabytes) over one month of logging," according to the researchers. The dataset which was collected in June 2006 contains summaries of 30 billion conversations among 240 million people. And they were very surprised to find that the average number of jumps to get from one random user to another was 6.6." This is very close to the old 'six degrees of separation' idea which states that everyone on Earth is six 'steps' away from anyone else. But read more...
You can see above a map showing the geographic characteristics of worldwide communication. "Each point is a location of a user login obtained from a mapping of IP address to geographical coordinates. The color of the dot corresponds to the number of logins from that particular location." "Cold blue colors correspond to low values, green to yellow colors to medium values, and hot red colors correspond to high values in the plots." (Credit: Microsoft Research)
This second map shows "the number of users at particular geographic location superimposed on the map of the world. Color represents the number of users, and blue circles represent locations with more than 1 million users." (Credit: Microsoft Research) As you can see, the northern hemosphere is largely over represented.
This study has been led at Microsoft Research Redmond lab by Eric Horvitz and Jure Leskovec, who was an intern at the time. "Horvitz says he was surprised that their analysis so closely matched the 1967 result. He wonders whether the number six is a basic constant for social interactions. 'Do we have a natural harmonic for social communication?' he asks. 'This is my conjecture -- more work needs to be done on that.'"
Of course, as you can see in the second map above, "there are obvious biases in the data, with 15- to 30-year-olds being by far the biggest groups of users. Geographically, the majority of users were in North America, Europe and Japan; large areas of the developing world provided no data at all. In Africa, most users were located around coastal areas, and North Korea was completely 'dark.'" And obviously, this study covered only users of Microsoft messenging system.
This research work has been published by Microsoft under the name "Worldwide Buzz: Planetary-Scale Views on an Instant-Messaging Network." Here are links to two slightly different versions of the abstract, one from Microsoft, the other one posted on arXiv.
Here is the second version. "We present a study of anonymized data capturing a month of high-level communication activities within the whole of the Microsoft Messenger instant-messaging system. We examine characteristics and patterns that emerge from the collective dynamics of large numbers of people, rather than the actions and characteristics of individuals. The dataset contains summary properties of 30 billion conversations among 240 million people. From the data, we construct a communication graph with 180 million nodes and 1.3 billion undirected edges, creating the largest social network constructed and analyzed to date. We report on multiple aspects of the dataset and synthesized graph. We find that the graph is well-connected and robust to node removal. We investigate on a planetary-scale the oft-cited report that people are separated by ``six degrees of separation'' and find that the average path length among Messenger users is 6.6. We also find that people tend to communicate more with each other when they have similar age, language, and location, and that cross-gender conversations are both more frequent and of longer duration than conversations with the same gender."
And here are links to two versions of the technical paper, one from Microsoft (PDF format, 37 pages, 1.50 MB), the other one posted on arXiv (PDF format, 26 pages, 1.37 MB). Both versions contain the illustrations featured above.
The results of this study will be presented at WWW2008, the 17th International World Wide Web Conference which will be held on April 20 to 25, 2008 in Beijing, China.
And don't think that Horvitz is just resting and waiting for this conference. In its March/April 2008 issue, Technology Review has published a series of articles about the 10 Emerging Technologies of 2008. One of them is Modeling Surprise. "Combining massive quantities of data, insights into human psychology, and machine learning can help manage surprising events, says Eric Horvitz." Busy man...
Sources: Katharine Sanderson, Nature, March 13, 2008; and various websites
You'll find related stories by following the links below.