Dr Adam Dennett presented the 2017 prize for the best paper on spatial analysis, in memory of Sinesio Alves Junior, to Luis Francisco Mejia Garcia for his paper entitled: “Modelling Spatial Behaviour in Music Festivals Using Mobile Generated Data and Machine Learning” (with co-authors Guy Lansley and Ben Calnan).
CASA has presented this award annually at GISRUK since 2010 in memory of Sinesio Alves Junior (pictured above). Junior, as he was affectionately known, is a much missed friend who joined CASA in 2001 as a Doctoral Researcher and member of staff.
Previous winners of this prestigious GISRUK award have been:
Kaisa Lahtinen for “Using Geographically Weighted Regression to Explore Spatial Variation in Survey Behaviour” in 2016 held at the University of Greenwich
Alexandros Alexiou for “The Role of Geographical Context in Building Geodemographic Classifications” in 2015 held at the University of Leeds
Sarah Wise for “Agent Based Modelling and GIS: Exploring our New Tools in a Disaster Context” in 2014 held at the University of Glasgow
Dan Olner, Alison Heppenstall and Gordon Mitchell for “Modelling and visualising the impacts of increasing fuel costs on the UK spatial economy in 2013 held at the University of Liverpool
Suzanne Mavoa, Melody Oliver, Nicola Tavae and Karen Witten for “Using GIS to integrate children’s walking interview data and objectively physical activity data” in 2012 held at Lancaster University
Heshan Du, Suchith Anand, Mike Jackson, Didier Leibovici, Jeremy Morley and Glen Hart for “Geographical Information Integration from Disparate Sources” in 2011 held at Portsmouth University and
Didier Leibovici, Lucy Bastin, Suchith Anand, Jerry Swan, Gobe Hobona and Mike Jackson for “Spatially Clustered Associations in Health GIS” in 2010 held at UCL.
Below is a summary of the 2017 winning paper and you can access the full paper by clicking the link below:
This study explores the utility of location data collected from a mobile phone app as a means of modelling spatial behaviour for consumer analysis, focusing on data from a music festival. Our aim was to harvest geo-temporal variables from the app data to model when individuals visit catering services across the site. Using Random Forest and Artificial Neural Networks machine learning algorithms, we presented an efficient means of simulating the popularity of bar areas within the festival site across time. The research demonstrates that with an appropriate methodology, mobile app data can provide useful insight for service provision planning.