Tuesday, 1 July 2014

GIS and data mash-ups

Xiyu Phoon - Consultant
It has been estimated that around 90 per cent of data today was churned out in the last two years. With the explosion in social media and provision of open government data, perhaps it comes as no surprise that the era of ’Big Data’ has arrived.
Every morsel of data, no matter how insignificant it may be perceived to be, has a story to tell. With almost unfettered access an immense library of data at our fingertips, we can analyse and visualise data in all sorts of ways and find some compelling stories.
As an example, we might be interested to see whether there is a relationship between fast food and crime. We can download data on the number of fast food outlets per local authority from Public Health England and crime deprivation from the Open Data Communities website. The number of fast-food outlets is given as a rate per 100,000 of the population, whereas crime deprivation includes burglaries, thefts, criminal damage and violence and is ranked from most to least deprived based on population-weighted average ranks of lower super output areas within local authorities.
When we plot the values together, we see a negative trend, suggesting that the most crime-prone areas have the most fast food outlets. However, we cannot discern from these datasets alone whether this is because fast food promotes criminal behaviour or that fast food outlets tend to pop up in the most deprived areas, which happen to have higher crime rates. As with all forms of data analysis, there will undoubtedly be some caveats, such as how data sets were collected in the first place and whether they are comparable.
Given that most data has an inherent spatial component, we can take such basic data analysis further and plot a map, which can be quite useful to see the distribution of fast food outlets.
As we can see, there is plenty of data available from a number of different sources which can be combined in order to discover all sorts of interesting titbits of information which can be presented in a variety of ways.
Our day-to-day projects at Temple focus primarily on environmental data, but we are increasingly making use of open data to enhance our analyses and help our clients with more in-depth decision-making.

No comments:

Post a Comment