PhD Coffee Mornings

This page contains some short minutes from our weekly PhD meetings. Blog entries will have fuller information.

Monday 15th June

Nawaf began the meeting by presenting his recent work on spatio-temporal patterns of motor vehicle theft (MVT) in Riyadh using Crime Analysis. In particular, he used the raster calculator to quantify changes at different time periods. See this blog post for more info. Nick Hood recommended looking at Flow-Co maps. Nawaf agreed to write a short post for this work.

Flow-co maps:

• The author Nick mentioned: Jonathan Corcoran
• A paper: http://0-www.tandfonline.com.wam.leeds.ac.uk/doi/pdf/10.1111/j.1467-9272.2007.00639.x
• Another: http://ac.els-cdn.com/S0198971506000354/1-s2.0-S0198971506000354-main.pdf?_tid=ae7ac590-1363-11e5-9e98-00000aacb35e&acdnat=1434375635_4b295964a8f8946d2382eb43dfeedf64

Rui asked about how to model the change in a vegitation index. The problem is finding a model similar to GWR that also takes a temporal element into account. Nawaf suggested a Bayesian spatio-temporal model. It’s an interesting problem I’m sure we’ll come back to.

Eusebio finished the meeting by presenting his ongoing work using machine learning methods for predicting passenger flow on railways. There is no real data yet, but it’ll be interesting to see how these work on a real system.

(Minutes by Nick M)