Nature mag­a­zine cover story dis­cusses rev­o­lu­tionary findings

In a ground­breaking paper pub­lished as a cover story in this week’s Nature mag­a­zine, North­eastern Uni­ver­sity physi­cist Pro­fessor Albert-​​László Barabási and his team found that humans can be char­ac­ter­ized based on how they move. In the article, titled “Under­standing Indi­vidual Human Mobility Pat­terns,” the authors dis­cuss how, for the first time, they were able to follow indi­vid­uals in real-​​time and dis­cov­ered that despite the diver­sity of their travel his­tory, humans follow simple repro­ducible patterns.

Barabási, along with co-​​authors Marta C. González and César A. Hidalgo, studied the tra­jec­tory of 100,000 anonymized cell phone users – ran­domly selected from more than 6 mil­lion users – and tracked them for a six-​​month period. They found that con­trary to what the pre­vailing Lévy flight and random walk models sug­gest, human tra­jec­to­ries show that while most indi­vid­uals travel only short dis­tances and a few reg­u­larly move over hun­dreds of miles, they all follow a simple pat­tern regard­less of time and dis­tance, and they have a strong ten­dency to return to loca­tions they vis­ited before.

We found that human tra­jec­to­ries show a high degree of tem­poral and spa­tial reg­u­larity, each indi­vidual being char­ac­ter­ized by a time-​​independent char­ac­ter­istic travel dis­tance and a sig­nif­i­cant prob­a­bility to return to a few highly fre­quented loca­tions, like home and work” said Albert-​​László Barabási, Dis­tin­guished Pro­fessor of Physics and Director of the Center for Com­plex Net­work Research (CCNR) at North­eastern University.

Our study shows that humans, after only three months of sat­u­rated behavior, reach sta­bility in their mobility pat­terns, and the tra­jec­to­ries become iden­tical,” added Marta C. González, Ph.D. in Physics and Research Assis­tant at the CCNR. “People devote their time to a few loca­tions, although spending their remaining time in five to 50 places, vis­ited with dimin­ished regularity.”

The loca­tion of cell phone users was located every time they received or ini­ti­ated a call or a text mes­sage, allowing Barabási and his team to recon­struct the user’s time-​​resolved tra­jec­tory. In order to make sure that the find­ings were not affected by an irreg­ular call pat­tern, the researchers also studied the data set that cap­tured the loca­tion of 206 cell phone users, recorded every two hours for an entire week. The two data sets showed sim­ilar results, the second val­i­dating the first.

The find­ings of this research com­ple­ment the notion that human mobility can be gen­er­al­ized by the Lévy flight sta­tis­tics, as sug­gested by a 2006 study that found that bank note dis­persal is a proxy for human move­ment. That study ana­lyzed the dis­persal of about half-​​a-​​million dollar bills in the U.S. and con­cluded that human travel on geo­graph­ical scales is an ambiva­lent and effec­tively superdif­fu­sive process. By using a dif­ferent method­ology, Barabási’s group was able to find evi­dence to com­ple­ment those findings.

Con­trary to bank notes, mobile phones are car­ried by the same indi­vidual during his/​her daily rou­tine, offering the best proxy to cap­ture indi­vidual human tra­jec­to­ries, said César A. Hidalgo, Ph.D. and Research Assis­tant at the CCNR. “Also, unlike dollar bills that always follow the tra­jec­tory of the cur­rent owner and dif­fuse, humans dis­play sig­nif­i­cant reg­u­larity and do not diffuse.”

The inherent sim­i­larity in travel pat­terns of indi­vid­uals could impact all phe­nomena driven by human mobility, from epi­demic pre­ven­tion to emer­gency response, urban plan­ning, traffic fore­casting and agent-​​based mod­eling,” added Barabási.

The study relied on a sample from anonymized, aggre­gate billing data from cell-​​phone users in an uniden­ti­fied Euro­pean country. The Insti­tu­tional Review Board at the U.S. Office of Naval Research, which funded this study as part of a larger pool of research into human mobility pat­terns, reviewed the pro­posal in June 2007 and deter­mined that it did not involve human subjects.

For more infor­ma­tion, please con­tact Renata Nyul at 617–373-7424 or at r.​nyul@​neu.​edu.

About North­eastern

Founded in 1898, North­eastern Uni­ver­sity is a pri­vate research uni­ver­sity located in the heart of Boston. North­eastern is a leader in inter­dis­ci­pli­nary research, urban engage­ment, and the inte­gra­tion of class­room learning with real-​​world expe­ri­ence. The university’s dis­tinc­tive coop­er­a­tive edu­ca­tion pro­gram, where stu­dents alter­nate semes­ters of full-​​time study with semes­ters of paid work in fields rel­e­vant to their pro­fes­sional inter­ests and major, is one of the largest and most inno­v­a­tive in the world. The Uni­ver­sity offers a com­pre­hen­sive range of under­grad­uate and grad­uate pro­grams leading to degrees through the doc­torate in six under­grad­uate col­leges, eight grad­uate schools, and two part-​​time divi­sions. For more infor­ma­tion, please visit www​.north​eastern​.edu.