Last month we welcomed five members of the LIDA Data Scientist Internship Programme to the Centre. Over the next six months George Breckenridge, Stuart Ross, Sebastian Heslin-Rees, Rosalind Martin and Simon Leech will be working with CDRC researchers on the following projects:
Analysing COVID-19 Mobility Responses through Passively Collected App Data – George Breckenridge and Stuart Ross
‘Lockdown’ policies restricting mobility have caused mass disruption to the normal operation of daily activity in cities across the COVID-19 pandemic. They mark the first time in recent memory that national and global populations have been caused to simultaneously re-evaluate transport choices, whilst also causing wholesale changes in the location and spatial footprint of most social and economic activity. The understanding of such dislocated geographies will underpin urban and transport planning policies for maintaining low virus transmission risk and for revitalising the UK economy, far into 2021 and beyond.
The emergence of passively collected anonymous mobility datasets, produced though mobile phone apps in cases where user permission is granted, makes it possible to explore these transportation responses at fine spatial and temporal scales – before, during and after the COVID-19 lockdown(s). The aggregation of these contemporary UK patterns – which will be required to maintain user anonymity – allows for the exploration of hundreds of thousands of users whilst simultaneously protecting privacy. Indeed, the utility of privacy-enhanced outputs for policy will be a lead project focus. Phone data will be provided by partner Cuebiq through their secure online platform, which enables only the export of aggregated outputs to suitable spatial units.
In order to investigate such unprecedented changes in mobility using Cuebiq’s data, we expect to employ a variety of machine learning (ML) methods to extract features. A journal paper documenting these patterns as the COVID-19 crisis evolves is the anticipated output for this CDRC-funded project
New insights into workplace and retail dynamics for English and Welsh cities – Sebastian Heslin-Rees
This project will be using Whythawk data on commercial properties in England and Wales, at Lower Super Output Area. It will make use of existing methodologies applied to a different scenario to produce new insight on commercial property rent and spatial location.
A commercial geodemographic classification of workplace zones
This research endeavour will utilise the newly available Whythawk dataset to construct a model for presenting and thus, understanding the spatial distributions of workers and workplaces across England and Wales. Largely, this will involve clustering workplaces of similar characteristics to distil a set of key workplace types, which can subsequently be mapped and analysed. In addition, the dataset has made available details of workplaces that have not been present in previous workplace datasets, such as distinguishing different workplace functions within multi-level building complexes. Consequently, this could provide additional insights and novel avenues for academic research and policy initiatives.
Predicting commercial rents using novel Machine Learning approaches
Using novel big data, this study will assess mass market appraisal within the English and Welsh commercial rental market. Mass market appraisal is the valuation of properties at a given time, and is required to ensure each property makes the appropriate tax contribution. This study will use a large volume of data on commercial business type, rental and rateable values and numerous external environmental variables. A range of machine learning algorithms will be used to predict and appraise the commercial rental market in England and Wales.
The outputs will include academic papers focused on methodologies employed, CDRC datasets and detailed maps. These projects are expected to help property professionals better understand commercial rental pricing and businesses who use and occupy these spaces and also researchers who are interested in how property values interact with other aspects of the environment.
Isolation and Exclusion in a social distancing COVID world – Rosalind Martin
The disparate impacts of COVID-19 and the associated lockdown have been much discussed recently, particularly in terms of age, deprivation, or employment sector. As the UK and other countries emerge from quarantine, it is equally apparent that the after-effects are likely to be long-lasting, whether through continued mitigation efforts such as social distancing or the economic impacts of economic shutdown, and that these after-effects are likely to further unevenly impact some groups over others. There are many dashboards reporting information on COVID cases and deaths, but information on the impacts on general population and businesses is missing.
Our main objective is to advance understanding of the social and spatial impacts of emergence from lockdown, identifying those households and places at risk of further isolation, under a scenario of continued social distancing, high unemployment, and a potential contraction of local service provision, including public transport. We have three research questions:
1. Are some typical household structures more vulnerable than others as a result of social distancing (and what are they vulnerable to, e.g., unemployment, social isolation, decreased service provision versus decreased access to existing services, decreased mobility, etc.)?
2. Is there a critical intersection of mobility, employment status and social distancing rules that predispose households with particular structures to isolation?
3. Based on current neighbourhood patterns and planning, what is the geography of isolation vulnerability?
It is expected that each of these scenarios have a particular geography. The creation of this dashboard should help predict where geographies of isolation under intersecting scenarios occur. Identifying areas at risk of isolation and exclusion through this project could prove invaluable to local councils who will be working to ensure all individuals are given access to relevant levels of assistance and resources during COVID-19 recovery, rather than allowing pre-existing disparities to widen.
