Sourced and adapted from http://www.news.cornell.edu/stories/Dec10/UKmap.html (Dec. 20, 2010)
An Analysis of phone calls shows how regional boundaries could be ideally drawn by generating a map of regions of Great Britain in which people communicate more with each other than with outsiders.
Crowd sourcing technologies might change the way we view our boundaries , As Heidegger once said, a boundary is where reality begins it own unfolding ..
An interesting study coordinated between Cornell, the Massachusetts Institute of Technology and in the United Kingdom Analysed telephone calls in UK and generated a map in which the community/communication boundaries and patches were compared to boundaries between admistrative regions . The map is a result of looking at a database of 12 billion calls over a one-month period, from which 20.8 million nodes and 85.8 million links between them were extracted. The map of the island was divided into 3,042 pixels, each 9.5 kilometres square- The main measure used was total call time between nodes in each pixel and those in every other pixel. The computer then automatically generated groupings among pixels until it found an arrangement that showed the most links within communities and the fewest between them.
"A community is something where most of the links are inside the community instead of outside," explained It was Steven Strogatz, the Jacob Gould Schurman Professor of Applied Mathematics, that advised the researchers on how to analyze the data using an algorithm- The basis of the algorithm is on the number of links generated by phone calls inside a certain boundary – it is an interesting study since anthropology, sociology, mathematics and information technology experts all worked together to test our main hypothesis of a community – which is by definition an introverted group of people that share together values – mathematics interprets the quantity of the links generated by these values. IT will be more insightful to see if specific data mining on specific key words that relate to specific communal historical values can lead to the same results. .*
The research according to its instigator Dr, Strogatz saus that it could be applied in areas where conflict of boundaries cannot be resolved through technical ways such as surveying> this method seems to have a more basis into the social dimension as it relates directly to how people interact with each other and in that case it has a lot of capabilities if a historical dimension is also investigated.
Social science usually used observation – such as field trips and questionnaire in order to describe, analyse and study specific cultural and political organisations. It is here the overlay of geography that makes this research multi dimensional
The result is that the groupings coincided nicely with existing boundaries of administrative regions. The researchers came to the conclusion that "cohesive patterns within society promoting change in administrative boundaries and the latter, in turn, affecting human interaction."
Two main discrepancies from administrative boundaries were observed:
- People in eastern Wales communicated extensively with those around nearby cities in the adjacent region of West Midlands.
- Scotland communicated less with other parts of Britain.
- new "region" west of London was showing which is very close to a growing centre of high-tech industry
The methods used in this research can be extended to cell phone records, credit cards or personal travel patterns which would measure more personal as opposed to business communication, or with credit card transactions or personal travel patterns,
The research will be very interesting since it is giving a new social dimension to what preciously known as geographical analysis – it is the overlap of the geography with social that will give transportation planners, policy makers, local councils and other city managers the possibility to see linkages on their blind spot- is not this what planning is ? Is not this one of the way we can use technology to assist us in sustainable planning?
- The algorithm was developed by former Cornell researchers Michelle Girvan, now a University of Maryland physics professor, and Mark Newman, now a University of Michigan physics professor. The research is reported in the Dec. 8 issue of the online journal PLoS ONE. The Cornell algorithm is described in the Proceedings of the National Academy of Sciences 103: 89577-8582 (2006).