Description
This module introduces theories and techniques to visualise, model and analyse (big) spatio-temporal data. Students will be introduced to the topics of statistical modelling, data mining and machine learning, and will learn tools and techniques for spatio-temporal analysis, with an emphasis on application to real world problems. The module content covers a range of topics, which include: Exploratory spatio-temporal visualisation, Statistical modelling and forecasting, Clustering and outlier detection , Machine learning techniques (e.g. Support Vector Machines, Random Forests, Artificial Neural Networks and Deep Learning), Space-time multi-agent simulation, and Social media analysis. Lectures are supported by practical sessions, where real data is used to demonstrate the techniques, with applications such as environment, transport, crime and social media analysis. The software packages used include R (http://www.r-project.org/), SaTScan (http://www.satscan.org/), Python and NetLogo (https://ccl.northwestern.edu/netlogo/). The course is suitable for MSc students in GIS, Geospatial Analysis, Spatio-Temporal Analytics, Smart Cities, Computer Science and related subjects.
Learning Outcomes
Understand the basic principles and techniques of spatio-temporal analysis and modelling
Be comfortable working with spatio-temporal data of different types in different application areas
Be familiar with using R statistical package for space-time analysis, modelling and visualisation
Have a working knowledge of other software such as SaTScan and NetLogo.
Be able to apply the tools and techniques they have learned to new datasets.
