Thoughts on Research Papers
The Livehoods Project: Utilizing Social Media to Understand the Dynamics of a City
Cranshaw et al. (2012)
This is definitely one of the most interesting papers I have come across. In conceptualizing and exploring the city we rely on a range of smaller areas — neighborhoods, boroughs, wards, and districts — in order to make urban space intelligible. While we can readily discuss how neighborhoods are shaped by physical geography (topography, adjacency to lakes or rivers, etc.), ordinance (zoning, access to public transit) and economics (real estate prices, average resident income), machine learning does not really spring to mind when we are considering how we might define ‘a neighborhood’. Livehoods is a project hatched within the School of Computer Science at Carnegie Mellon University that leverages 18 million Foursquare check-ins to draft up new urban ‘activity zones’ based on the patterns of frequent visitors. The venture essentially asks how does a location-based service reflects our sense of place within the city?
The paper shows how beautifully we can visualize and understand the dynamics of a city using data. Most of the people living in a particular city know which part of it is rich and which are underdeveloped. We take this for granted even though our shopping behavior, hang out places, etc are governed by this “inner feel” we have for each locality. Building a ‘whole foods’ store had a huge economic and cultural impact in East Liberty and made me realize how such stores impacted my behavior in the past in my home city. How municipal corporations at times arbitrary and some times calculated building of infrastructures like Metro or Roads can bring about a change in the behavior of people living there and how using data can be useful in this regard. If they could have made their data even less biased by using the data available directly from FourSquare API instead of data only shared on twitter would have been even better. Municipal corporations and governments should use such data and make judicious decisions while allocating budgets/funds instead of using ‘set in stone’ boundaries they use. The use of spectrum clustering technique along with Affinity matrix to blend spatial affinity and social affinity was novel. The visualization tool provided was beautiful.
This is one of those papers for me where you didn’t know data can have such a huge impact in a domain and the field of urban computing is interesting. It certainly shows how mental and physical boundaries define our thinking and how it can be changed by the use of data.
