Home » Archives for Robyn Abbott

GISRUK 2020 – CDRC Researchers

GISRUK 2020 – CDRC Researchers

Earlier this month CDRC researchers and students from the CDT Data Analytics and Society presented papers (virtually) as part of the 28th Geographical Information Science Research UK Conference.

Measuring Lifestyle Courier Work Area Similarity – Kostas Cheliotis, Sarah Wise, Fraser McLeod, Tom Cherrett, Julian Allen, Oliver Bates, Maja Piecyk and Tolga Bektas. 

High-Frequency and High-Resolution Neighbourhood Housing Price Dynamics: Identifying Spatio-Temporal Hotspots and At-Risk Areas for London, England – Jacob L. Macdonald. 

Evaluating the impact a restaurant aggregator might have on a UK National Restaurant Chain and with that impact in mind consider whether prevailing retail theories apply in the online world – Jason Dalrymple.

Estimation of small-area tourist demand using Airbnb accommodation in London – Zi Ye, Graham Clarke and Andy Newing. 

Must all maps be the same? A Mixed Quantitative-Qualitative Approach to Resource Allocation – Andrew Renninger. 

Towards data-driven human mobility analysis – Terje Trasberg and James Cheshire. 

Measuring Beauty in Urban Settings – Alessia Calafiore. Measuring Beauty in Urban Settings.

A Global Comparison of Bicycle Sharing Systems – James Todd, Oliver O’Brien and James Cheshire. 

Proposing a new method for creating integer weights for Iterative Proportional Fittings based on locally calibrated model – Jason Chi Sing Tang and James Cheshire. 

Retail Vibrancy and the Composition of the UK High Streets – Abigail Hill and James Cheshire. 

Towards Real-time Agent-Based Pedestrian Simulation using the Ensemble Kalman Filter – Keiran Suchak, Nick Malleson, Jon Ward and Minh Kieu

GISRUK is the largest academic conference in Geographic Information Science in the UK. Since 1993, GISRUK has attracted international researchers and practitioners in GIS and cognate fields, including geography, computer science, data science, and urban planning to share the latest advances in spatial computing and analysis.

COVID 19 – Providing new insights to aid societal recovery

Birds eye view of a crowd of people on street

COVID 19 – Providing new insights to aid societal recovery

Earlier this year we announced our involvement in the Emergent Alliance – a not-for-profit community established to share data, expertise and resources to work together to aid economic recovery in 2020 and shape a new normal.

We are working through our parent organisation Leeds Institute for Data Analytics to provide the COVID-19 data alliance with scientific expertise and access to global academic research networks.

What is the Emergent Alliance?

The Emergent Alliance is a collaboration of large organisations, small businesses, institutes and individuals – founding members include Leeds Institute for Data Analytics, IBM, Google Cloud, The Data City, Truata, Rolls-Royce, Microsoft, ODI Leeds, SATAVIA, Fieldfisher and Whitespace.  Since launching in April we have added a further 29 members and our Alliance continues to grow globally.

Drawing on this diverse collaboration of corporates, individuals, NGOs and Governments, the Alliance will contribute expertise, data, and resources to inform decision making on regional and global economic challenges to aid societal recovery post Covid-19.

We want to have the ability to take a broad set of economic, behavioural and sentiment data, fuse it together and provide new insights and practical applications to the global Covid-19 response.

The models produced will identify lead indicators signalling economic recovery cycles that global businesses and government can use to build operating confidence in investment and activities that shorten or limit recessionary impacts.

Who and how can the Alliance help?

Governments, businesses and individuals around the world have been challenged by Covid-19 to act quickly, decisively and effectively using the best available scientific evidence and insight.

The Alliance provides a much-needed independent means of sourcing and shaping ideas regarding Covid-19 recovery and longer-term sustainability is required.

By mobilising skills, data and analytics from a wide range of organisations and institutions at a global scale, the Emergent Alliance can help to contribute to a step-change in the use of data to address the economic and social challenges which come in its wake.

The outputs from the alliance are intended to help understand economic implications and aid recovery, to provide insight beyond what is currently known.

Our role in the Alliance

LIDA has a number of roles within the Alliance, including:

  • providing a secure, accredited and independent data sharing platform;
  • offering data analytics support for the execution of projects;
  • leading the preparation of ‘challenge statements’;
  • and it is one of a small number of members with responsibility for shaping the alliance’s strategic direction.

CDRC is working through Leeds Institute for Data Analytics to provide the Alliance with scientific expertise and access to global academic research networks.

Current Challenges

Activities are structured around a series of Challenge Statements, which articulate the social and economic problems to be addressed. The Alliance currently has four challenges underway:

As well as sharing results from challenges, members are asked to make their data available publicly wherever possible. There has been a lot of progress in curating data sets and sharing information through an online cataloguedeveloped by Open Data Institute (ODI) Leeds.

Whilst the Emergent Alliance does focus on the recovery from Covid-19, its cross-industry collaboration model and practices for data sharing could be applied to any number of pre-existing societal challenges.

Ads for junk food in the UK seem to be concentrated in poorer areas

Burgers and chips

Ads for junk food in the UK seem to be concentrated in poorer areas

Billboards advertising unhealthy food are concentrated in poorer areas and areas with a higher proportion of overweight children in Liverpool, UK. These findings may also apply across the country.

Using a combination of artificial intelligence and street-view images, CDRC researcher Mark Green and his colleagues at the University of Liverpool mapped the content and geographical location of more than 10,000 outdoor adverts in the city.

The research featured in New Scientist this week, where subscribers can read the article in full.

Dr Mark Green

Dr Mark Green is a Senior Lecturer in Health Geography at the Geographic Data Science Lab at the University of Liverpool.

Geographical determinants of health

Mark’s research explores the ways in which features of the neighbourhoods we live and interact with daily imprint on our health. His previous work includes:

Find out more about Mark’s research and the Geographic Data Science Lab at the University of Liverpool.

Championing Localism – No more road to nowhere

Better towns roadmap road sign

Championing Localism – No more road to nowhere

Innovative roadmap launched to lead the way towards better, healthier and more successful towns

Today sees the launch of a new and unique approach to support towns and the challenging journeys they face. The Better Towns Roadmap consortium has created a highly visual and interactive roadmap to help transform towns across the UK. The consortium brings experienced practitioners together with academics who share a passion for repurposing towns, rejuvenating forgotten spaces and engaging their communities.

A collaboration between HLM Architects, Didobi and realestateworks has developed a step-by-step roadmap to realise any town’s short or longer term goals. It creates clear links between a town’s vision, goals, and targeted outcomes, presenting customised outcomes that defy traditional ‘one size fits all’ approaches to regeneration, adaptation and change.

The decision to create #bettertowns was driven in part by frustration at the lack of effective collaboration across the industry and the desire to join up projects from their conception to delivery.

The belief is that multi-disciplinary, collaborative, data-driven approaches can transform any town’s prospects through a succession of deliverable projects linked by a unique vision.

The #bettertowns roadmap facilitates the successful repurposing of towns where rigour, process and logic are applied at every stage.
Each town’s journey proceeds by successively creating a baseline, defining a mission, appraising options, creating an action plan and delivering outcomes. Each step is further unpacked on the bettertowns website and is supported by an extensive virtual library and self-assessment questionnaires.

Olivia Paine, HLM Architect’s Asset & Workplace Project Lead, has worked with many local authorities to help rationalise and revitalise their assets. She is passionate about the collaborative potential of #bettertowns which will help local authorities not only to achieve goals but to understand the process behind the journey:

“Towns have faced increasing challenges over the years and the current pandemic has seen a shift in the pace of change and an increased necessity for well-informed, connected local authorities. We understand that every town has the potential to be unique and welcome the return of ‘localism’ to the agenda. It is not about selling a generic product, it is about sharing knowledge and information
to support towns achieving their desired outcomes.”

Brian Thompson, founding Director of realestateworks, a niche consultancy specialising in public sector and collaborative asset management, warns against quick fixes to systemic issues facing the sector:

“Every town is infuenced by but also shapes the economy of neighbouring towns and communities. We look beyond quick fixes and beyond simply the High Street for only by doing so can sustainable, viable, and long-lasting strategies be defined.”

Matthew Hopkinson, Co-Founder and Director at Didobi, is a well known and highly respected practitioner in this space having been involved in a number of High Street Reviews as well as having created data analytics, aide-memoires and research on towns:

“I am delighted that after a year of planning we are able to launch the Better Towns Roadmap. I believe that this highly visual, clearly staged and evidence driven approach is what our towns need in order to address the challenges they face today and in the future. It is a long journey and one that needs to be measured against key milestones if success is to be realised and I hope that by sharing the Better Towns Road Map approach we can support all towns on their journeys.”

A partnership with the Consumer Data Research Centre (CDRC) brings additional benefits to towns and their communities by providing access to additional insights, tailored research, unique datasets and intelligence from the academic domain.

Professor Paul Longley, Director and Principal Investigator at the CDRC welcomes the opportunity to support the repurposing of towns using unique and timely analysis ready data:

“The CDRC is delighted to contribute to the Better Towns Roadmap consortium and strengthen academic links with businesses and local authorities, particularly at this challenging time. We believe that we bring in-depth expertise in data and analysis to this important initiative.”

The consortium will be crowdsourcing data via a self-assessment page on the website. Data will be captured from representatives of councils, businesses and communities on their towns in order to help understand the issues and opportunities that everybody faces. The data will be aggregated, analysed, anonymised and shared in order to better understand different perceptions of salient shared issues.

A frequently updated key publications library will be maintained on the website. This will help raise the overall level of awareness of the tools,techniques, guidance and success factors among those that share interest in creating better towns.

Visit www.bettertowns.co.uk for more information on the roadmap and details of the self assessment

Notes to editors:

Created in 2020 the Better Towns Roadmap consortium was formed to work with towns, both large and small and to bring a multi disciplinary approach to re-purposing towns from the beginning of the journey through to the execution of a successful plan.

Based on the collective idea that multi-disciplinary, collaborative, data-driven approaches can transform any town’s prospects the Better Towns Roadmap consortium consists of a group of respected experts, that individually bring unique knowledge and expertise.

They understand towns from an occupier, investor, local authority, and community perspective, with first-hand experience, witnessing the benefits of combining architectural creativity, data-led insights, effective stakeholder engagement, and feasibility testing. They bring a deep knowledge of retail data, places and trends and are rigorous, independent, and objective in their approach, design, analysis, data and systems.


What can tweets about contact tracing apps tell us about attitudes towards data sharing for public health? (Part 3)

Crowd of people in railway station

What can tweets about contact tracing apps tell us about attitudes towards data sharing for public health? (Part 3)

At the end of my last blog post about Covid-19 apps, I speculated it was unlikely that the UK’s Track and Trace app would gain enough public trust and support to be a success.  Since that blogpost was published, it was announced that UK’s app might not be ready until winter, followed by news that the centralised NHSX app has been abandoned for a decentalised alternative developed by Apple/Google.  

Many people have reacted to this news on Twitter resulting in a spike of tweets about the Track and Trace app (Figure 1). In this blog post I will present findings from sentiment analysis on these tweets to understand people’s reactions to the new decentralised app and discuss the future of data-sharing post-Covid-19.  

Graph showing number of tweets about Covid-19 tracing apps
Figure 1: Daily number of tweets about all ‘Covid-19’/’Coronavirus’ apps from all countries (blue) and only  the UK’s ‘Track/Test and Trace’ app (orange), collected 24 April to 16 June 2020 and 17 June to 25 June 2020 respectively.  

Holly Clarke

Leeds Institute for Data Analytics

Holly Clarke is an Intern at Leeds Institute for Data Analytics, applying data science solutions to solve complex, real-world challenges. She is working for the LifeInfo project with Michelle Morris, researching attitudes towards novel lifestyle and health data linkages and how access to this information could improve public health. 

Read the previous parts of this blog:

Read part 1
Read part 2

The positives and negatives of Covid-19 apps  

This sentiment analysis includes tweets about the UK’s Track and Trace app posted between 17th and 25th June 2020, thereby, focusing in on recent events.  Sentiment analysis matches words within tweets with common positive and negative words categorised in the “Bing” dataset, developed by Bing Liu in order to identify their sentiment. Overall, this analysis tells us there are more commonly used negative words within recent tweets about Track and Trace app than positive, indicating the tweets hold mainly negative content (Figure 2).  

Proportion of sentiment words in tweets that are positive and negative for tweets about all Covid-19 apps, collected 24 April to 16 June 2020, and tweets about the UK’s Track/Test and Trace app, collected 16 June to 25 June 2020.
Figure 2: Proportion of sentiment words in tweets that are positive and negative for tweets about all Covid-19 apps, collected 24 April to 16 June 2020, and tweets about the UK’s Track/Test and Trace app, collected 16 June to 25 June 2020.  

The nature of these positive and negative words is also very telling. The negative words refer predominantly to the management of the app rather than issues about data privacy and the app itself; “failure”, “incompetence”, “fiasco”, “shambles”, “chaos”, “disaster”, “lying” and “debacle” all feature prominently (see Figure 3).  As a comparison, sentiment analysis on all general tweets from 24 April to 16 June 2020 about Covid-19 apps (Figure 4) shows common negative words to be more technology focused and in line with common concerns about data-sharing – “breach”, “risk”, “concerns”, “issues”.  

50 most frequently used positive and negative sentiment words used in tweets about the UK’s Track/Test and Trace app, collected 16 June to 25 June 2020.
Figure 3: 50 most frequently used positive and negative sentiment words used in tweets about the UK’s Track/Test and Trace app, collected 16 June to 25 June 2020.  

The positive words refer to more common topics around data-sharing and technology e.g. “trust”, “protection” and “safe” across both datasets of tweets. This indicates engagement with the topic of data sharing and a significant proportion of the tweet sentiment words are positive across both datasets.  

As part of the sentiment analysis I have controlled for negation, inversing the positive/negative categorisation if a common negator is directly before the word (e.g. “not good” or “don’t trust”). In the figures these are shown with the pre-fix “neg_”.  However, linguistic features such as sarcasm, humour and questioning are not easily picked up through sentiment analysis. Some instances of positive words like ‘wow’ or ‘promises’ may also be used in a critical way.   

Overall, although both datasets of tweets include more negative than positive words, the recent events around the UK’s Track/Test and Trace app seem to have framed the app more negatively than Covid-19 apps more generally due to “waste” and issues around the development of the app.   

50 most frequently used positive and negative sentiment words used in tweets about ‘Covid-19’/’Coronavirus’ apps from all countries, collected 24 April to 16 June 2020.
Figure 4: 50 most frequently used positive and negative sentiment words used in tweets about ‘Covid-19’/’Coronavirus’ apps from all countries, collected 24 April to 16 June 2020. 

What will Track and Trace mean for people’s attitudes to data sharing?  

When I began writing this blog series on Covid-19 apps, countries across the world were rapidly launching contact tracing apps to quelle the spread of coronavirus through technology and the UK was poised to trial their app on the Isle of Wight. Two months later the Track and Trace app journey has certainly not been smooth and the app’s importance has been downgraded from “world beating” to “the cherry on the cake”.  But what does this late-stage Apple/Google switch will mean for public opinion?  

Research on attitudes to data-sharing, as discussed in my last blog post, frequently finds that people’s willingness to share their data is dependent on which actors are involved. People tend to have high trust in the NHS and the lowest trust in private companies. Hence, we might expect the shift from an NHSX app to one involving tech giants Apple and Google to be met with opposition. However, the sentiment analysis indicates conversation about the Track and Trace app mainly focuses on the wastefulness and “shambles” of the switch rather than inherent mistrust in private companies.  

Initial findings from my work with the LifeInfo project, exploring public opinion about linking lifestyle data (e.g. supermarket loyalty card or fitness app data) with health records, may explain this. My analysis highlights that data-sharing and trust in actors is not as straight forward as might be expected.  

Although people generally have high levels of trust in health organisations, respondents repeatedly expressed concerns that their supermarket loyalty card data might be seen by their GP if these data were linked for health research. Many worried their GPs would unfairly judge their diet and lifestyle, and even withhold treatment. Yet, respondents were happy for supermarkets (private companies in which research finds people to have the least trust) to store and use their loyalty card data.  This indicates that attitudes about data sharing are not simply informed by trust in actors but are also influenced by the type of data involved and social norms about how it is currently used.  

In the context of coronavirus apps, this could mean that users are more comfortable with mobile phone providers using data to alert them about exposure to coronavirus than the government or NHS. Many mobile phone users share vast amounts of data with technology companies through everyday use of apps and services which they may be uncomfortable sharing with the government or healthcare providers.  Therefore, a contact tracing system involving Apple and Google, and especially a decentralised one which enables more data privacy, might encourage wider use than the NHSX app.  

The future of data sharing post Covid-19 

The Covid-19 pandemic will undoubtedly create lasting changing across many aspects of our lives including attitudes towards data sharing.  The pandemic had led us to consider sharing unprecedented amounts of data it has also made clear the inadequacies of our medical data sharing systems.   

In the context of the LifeInfo study, access to lifestyle data linked to health records could help researchers better understand and prevent diseases such as diabetes, certain cancers and heart disease, The World Health Organization attributes 30% of yearly global deaths to poor diet and physical inactivity, so it is a substantial challenge. However, for participants to willingly share their data they must trust organisations to safely, responsibly and transparently use it. 

Successful contact tracing apps had the potential to demonstrate that data sharing could help improve health while maintaining personal privacy and data security. Yet, technological failings, privacy concerns, and government mismanagement in the UK could turn public opinion against data sharing initiatives in the same way other high-profile failings such as care.data didAbove all, the Track and Trace app highlights how detailed consideration of peoples attitudes towards data sharing is vital for initiatives to be successful.   

Cities must act to secure the future of urban cycling: our research shows how

Cities must act to secure the future of urban cycling: University of Leeds research shows how


Robin Lovelace, University of Leeds (former CDRC researcher) and Joey Talbot, University of Leeds

Cities worldwide are preparing for the long transition out of lockdown. Physical distancing measures will be in place for many months, with impacts on all walks of life, not least transport.

With public transport options running at low capacity and emerging evidence of the role of air quality and exercise in mitigating the risks of COVID-19, solutions are needed more than ever.

New cycleways are being introduced in many cities, allowing healthy habits started during the lockdown to continue. Transport authorities must act fast, however, to take advantage of the current cycling boom while traffic levels are still below normal, and to avoid gridlock.

New cycleway on space reclaimed from motor traffic as part of the COVID-19 response in Park Lane, London.
Transport for London.


Our research at the University of Leeds has identified roads where there is both space and demand for cycling infrastructure.

Our methods have been used in a nationwide project funded by the Department for Transport and transport charity Sustrans to help relieve immediate pressures on the transport system and create long term change. The result is the Rapid Cycleway Prioritisation Tool. This is a free and open tool to help ensure that the government’s Emergency Active Travel Fund is spent where it is most needed, for maximum long term benefit.


New routes

The Rapid Cycleway Prioritisation Tool provides an interactive map for every transport authority in England. The main result is a list of road sections that have both high demand and sufficient space for cycling.

These are represented as blue lines in the map of Leeds below, many of which are included in the council’s new cycleway plans. Additional layers in the map include existing off-road cycleways (shown in green), which in many cities are disjointed or of variable quality, and, in purple, a vision of what a joined-up cycle network could look like.

The map output of the Rapid Cycleway Prioritisation Tool for West Yorkshire, zoomed in on Leeds.
University of Leeds

This new tool builds on our previous work, carried out with other universities, to help authorities identify and develop strategic cycle networks. We created a national dataset of roads, including estimates of the number of lanes in either direction and road width.

This can help identify where roads have space that could be reallocated to widen pavements or to rapidly introduce new cycleways. One of the first of these has been installed outside a hospital in Leicester. Community engagement through projects such as Widen My Path, which provides a forum for comments on where more space for walking and cycling is most needed, is another vital part of this process.

Read more:
Temporary urban solutions help us deal with crisis — and can lead to radical shifts in city space

The tool’s focus on road space reallocation is due to the importance of speed and capacity. Light segregation measures such as plastic bollards or wands can be implemented much faster than constructing new cycleways from scratch. Wide, direct and continuous cycleways, of the type created by road space reallocation on wide roads, are needed for capacity and to ensure physical distancing guidelines can be followed. Finally, road space reallocation has environmental benefits, representing “zero carbon infrastructure”.

An important finding is that many cities have wide roads with spare space, as shown the maps of six major cities below. Notably, none of these yet have a joined-up cycle network. An example of the type of road highlighted by our tool is Kirkstall Road in Leeds, which has high cycling potential and sufficient width for new cycleways. Kirkstall road is already part of plans by Leeds City Council to become a trial cycleway.

Our methods, based on open data and code, could be used in cities worldwide. Given the global nature of the challenges we face, open and collaborative research is vital. The potential for international application can be seen in research we carried out for the World Health Organization on possible cycling uptake in cities in low income countries.

Maps showing existing, disjointed cycleway networks (green), potential cycleway routes on wide roads according to the Rapid Cycleway Prioritisation Tool (blue) and cohesive networks (purple) in 6 major cities.
Robin Lovelace


Planning for the future

The UK government has announced £2 billion of investment in measures to promote walking and cycling in England over the next five years. £250 million has been allocated for emergency interventions to make cycling and walking safer.

Similar commitments are being made in Scotland and Wales. Local authorities urgently need to decide how this funding should be spent.

If action is prioritised in places where there is a long-term need for cycle improvements, there is a greater chance that these developments can become permanent. In Paris, COVID-19 related measures aim to contribute towards Mayor Anne Hidalgo’s long-term Plan Vélo.

Read more:
How major cities are trying to keep people walking and cycling

New cycling infrastructure is more likely to be effective when it is developed based on analysis of the best available data, in combination with vital local knowledge. City planners, politicians and citizens need to act to ensure that transport interventions made during the crisis are of maximum benefit now and in the post-pandemic world.The Conversation

Robin Lovelace, Associate Professor in Transport Data Science, University of Leeds and Joey Talbot, Research Fellow in Transport Data Science, University of Leeds

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Reflections on the impact of COVID-19 on grocery shopping

Reflections on the impact of COVID-19 on grocery shopping

Over the past few months we have all felt the impacts of COVID-19 on our grocery shopping. Be these direct impacts resulting from a government-imposed lockdown, changing of purchasing behaviour to reflect more time spent at home, or adapting to new financial realities. Grocery shopping is something we all need to do.

In the three weeks leading up to the UK lockdown on 22nd March, amidst the chaotic scenes inside supermarkets and in the news, supermarkets took an additional £1Billion per week compared to 2019. These revenues represent small increases in usual weekly household spend on certain products, with product shortages reflecting the ‘just in time’ basis of grocery supply chains where storing reserves of goods is limited.

Government advice to limit time spent outside the home is encouraging a return to the weekly shop. Consumers are making fewer trips for a higher quantity of goods. These trips need to accommodate for more time being spent at home as well as reduced opportunities to shop elsewhere. With the closure of bars and restaurants supermarkets are seeing an increased focus on meal planning.

Ryan Urquhart

Ryan is a PhD Researcher at the CDT – Data Analytics and Society, based at Leeds Institute for Data Analytics.

Front line retail staff have faced the brunt of these changes, accommodating optimised opening hours, social distancing measures and new staffing. New forms of demand segmentation are emerging, based not on traditional measures of loyalty and shopping mission, but to protect the most vulnerable and serve key workers. Government advice to switch to e-shopping is juxtaposed against retailers urging consumers to shop in store to save delivery slots for those who need them most. Despite widespread investment in delivery capacity and staffing (Tesco more than doubling weekly delivery slots to 1.2 million and Sainsbury’s pushing through a 75% increase to more than 600,000 slots), handling the increased demand is a substantial challenge – as anyone who has tried to book a delivery slot lately will have noticed.

Retailers have always needed to balance the need for home delivery coverage against profitability, with most home delivery services operating through the physical store network. Online orders are picked and packed from the shelves of supermarkets rather than through a separate distribution channel. Under normal circumstances, where 13% of the GB population regularly use home delivery services (Hood et al, 2020), this is an effective way to serve consumers and utilise existing supply infrastructure. However, when supermarkets are facing extreme upturns in demand, this system comes under pressure.

Access to home delivery services is not uniform across the country. Capacity is more readily available in areas more likely to have home delivery users under normal circumstances. These services have been built up over a period of time based on sophisticated analysis and location based decision making. My research seeks to address challenges relating to capacity management and home delivery provision. Taking an automated approach to geographic zone design we aim to divide the nation into self-contained areas for home delivery. Taking into account the underlying demand and available capacity, we aim to support the coverage, effectiveness and efficiency of home delivery services.

With online ordering set to continue after lockdown (a forecast 25.5% online growth in 2020), retailers need to adapt to serve a very different target demographic. The geography of e-commerce engagement is the subject of academic research. The CDRC Internet Users Classification shows how online engagement varies across the country. The most vulnerable in society, who are being prioritised for home delivery, will not uniformly live within areas associated with e-commerce usage. Thus retailers are facing a unique challenge in effectively prioritising vulnerable consumers, expanding services and serving a wider range of consumers whilst dealing with increased demand across the network.

Adapted from a University of Leeds, School of Geography Viewpoint piece.

Reflections on the dietary impacts of the COVID-19 lockdown period

Image of banana bread on chopping board

Reflections on the dietary impacts of the COVID-19 lockdown period

The closure of cafés, restaurants and work canteens during lockdown has imposed a significant change on the nation’s eating habits, with many of us now consuming all our meals at home. Time gained back from the daily commute has provided some with the welcome opportunity to reignite a love for home-cooking; my social media feeds have been filled with positive images of people finding pleasure in the wholesome comfort of nourishing the ones they love. But for others, the kitchen is uncomfortable territory and an added pressure at what is already a difficult time.

Undoubtedly, the changes in food availability, tighter purse strings and an increased focus on health have impacted our diets and relationships with food. Anecdotally, we hear of more people growing their own food, reducing food waste and making the effort to support local food suppliers. But exactly how the current situation has affected our diets and nutritional status is the topic of ongoing research, some of which you can contribute to here. My own PhD research explores the utility of supermarket transaction records as a measure of population diet; the primary data collection phase for which started just this week. I therefore have the additional challenge of acknowledging that data for the current period probably isn’t representative of the ‘norm’. But I also have a unique opportunity to contribute to understanding of how the nation’s diets adapted.

Vicki Jenneson


Vicki is a Nutrition PhD Researcher at the CDT – Data Analytics and Society, based at Leeds Institute for Data Analytics.

For some people, the lockdown has meant a struggle to put food on the table, with lost incomes and missed school meals seeing more families turn to Universal Credit and foodbanks for support. But we’ve also seen a tremendous community spirit, with innovative flexibility from businesses and local volunteers helping to bring food to the most vulnerable members of society.

Others have looked to nutrients as a means to ‘boost’ their immunity. Despite public health advice insisting that taking supplements cannot protect against COVID-19, a friend of mine working for a supplement manufacturer reports that she’s never been busier and is running out of national supplies. Indeed, that many nutrients have a role to play in the normal functioning of the immune response appears to contradict this message, highlighting just how difficult it is to communicate population dietary advice. A recent MyNutriWeb diet and immunity webinar series I attended highlighted the role of the ageing immune system (immunosenescence), and of our gut microflora in interacting with our immune response mechanisms (our guts contain ~70% of our immune cells). Although not exclusive to COVID-19, we may perhaps hypothesise that factors affecting our gut microbiota, may at least go a small way to explain some of the differences in response to the coronavirus that we’ve seen across the population. For example, older adults living in residential care homes have lower gut microbial diversity than those living in community, associated with poorer health outcomes. This may result from reduced dietary diversity and environmental exposures (including surfaces in the outside world and from the kitchen in which food is prepared). But, is it possible that some people’s diets may have actually improved during lockdown? Cooking at home and eating together as a family are both associated with a healthier diet. Increased meal planning has helped many of us reduce food waste, and fewer shopping trips are benefitting the environment too. Not only can home-cooking improve our relationship with food, it is associated with higher fruit and vegetable intake and better nutrient profiles.

For example, around 75% of the salt we consume is already present in the foods we buy, mostly processed foods and those eaten out. So, cooking from scratch gives us the opportunity to take control of our salt intake. Like all nutrients, salt does have an important role to play in our bodies; for regulating fluid balance, enabling nerve transmission and, for muscle contractions. However, consuming too much salt puts us at increased risk of cardiovascular disease, stroke and stomach cancer. Salt’s properties make it extremely useful for the food industry. Firstly, it adds flavour, and not just saltiness; a small amount of salt also enhances our perception of sweetness, without extra sugar. Secondly, it increases the shelf-life of foods by inhibiting bacterial growth. Finally, salt influences the texture of foods by interacting with proteins. While recent food industry reformulation has helped us reduce our salt intake from 9.5g to 8.1g per day between 2003 and 2011, efforts have been poorer in the out of home sector, which isn’t subjected to the same labelling rules as the retail sector. And, as a nation, we’re still eating more than the recommended maximum 6g (1 teaspoon) per day for adults and children over 11 years. But continuing to cook from scratch, replacing salt with herbs and spices, and making our own spice mixes, sauces and marinades can help us to further reduce our salt intake. As we cut down on salt, our tastes also change.

Our food choices and shopping habits are entwined with multiple aspects of social context, including affordability, car ownership, cooking skills, and cultural norms. And so, our ‘food experiences’ of the COVID-19 lockdown period will be vastly different, dependent on our situation, and likely to follow a socioeconomic divide which may indeed serve to widen dietary inequalities.

But just as we’re looking to emerge from lockdown with a more balanced and sustainable outlook on many aspects of our daily lives, whether that’s more flexible working hours, or a continuation of homeworking and the environmental benefits of reduced commuting, there is also potential for a fresh outlook when it comes to diet. What if we could harness the positive resilience that many have shown? What if the ‘dig for victory’ attitude, the community support networks, and the scratch-cooking revolution could indeed become some of the lasting legacies of the hardships of lockdown?

Adapted from a University of Leeds, School of Geography Viewpoint piece released on 1st June 2020.

COVID-19 Reflections: Using data to invent the future cities.

Empty traditional shopping arcade in Leeds

COVID-19 Reflections: Using data to invent the future cities.

The impacts of Covid19 on our daily lives and habitual routines are undeniable. These changes in our behaviour are reshaping our cities and urban areas; bustling hubs of productivity, creativity and innovation have been rendered temporarily dormant. Leeds is no exception to this, with the city centre experiencing an 89% decrease in footfall compared to the same period in 2019. Briggate is no longer experiencing an influx of Saturday morning shoppers and the once crowded bars on Call Lane are now deserted.

Despite the vast decrease in the footfall within Leeds city centre, it appears that people are still creatures of habit. The current patterns of hourly fluctuations in footfall counts remain similar to pre-pandemic Leeds. There is a rapid increase in footfall until the peak at noon, followed by a slight, gradual decrease in footfall until midnight. Tuesdays still experience the highest level of footfall, and Sundays the lowest. Changes in the night-time activity are evident, with footfall levels close to zero between midnight and 5 am.

Data can provide rich insights into urban dynamics and allow us to understand the complexity of cities. Without quantifying the current conditions of cities, it is challenging to identify how they should be developed and to assess any progress.
Pre-pandemic, the deployment of sensors to count pedestrian footfall, measure air quality and quantify noise levels were becoming a popular tool for local councils.

Annabel Whipp

Annabel is a PhD Researcher at the CDT – Data Analytics and Society, based at Leeds Institute for Data Analytics.

These data remain incredibly valuable, but now there is a drive for cities to utilise data to save lives and improve wellbeing. This can include the use of thermal imaging cameras to detect body temperatures and using drones to identify areas which require better management to allow for social distancing. There are understandable concerns about privacy and data protection. However, now more than ever, it is crucial to demonstrate the use of aggregated data for social good, while protecting individual privacy.

There are many questions as to what our post-pandemic cities will look like, given that the dynamics of everyday life will change. Professor Michael Batty notes the novel opportunity to invent our future cities, rather than to try to predict their complexity. We have the chance to redesign urban spaces with a new future in mind, one in which the health and wellbeing of citizens are at the forefront of urban design. We can use data to ensure future cities cater to the needs of citizens. Local councils and urban planners are already considering the ways urban areas can be ‘future-proofed’ and developed into safe and resilient cities. Plans are varied from pedestrianizing city centres and developing social-distancing friendly green spaces to increasing the role of health departments within urban design.

The future of cities may be uncertain for now. However, this pandemic has provoked discussion around what we want our cities to look like and how to redesign them.

Adapted from a University of Leeds, School of Geography Viewpoint piece

What can tweets about contact tracing apps tell us about attitudes towards data sharing for public health? (Part 2)

Crowd of people in railway station

What can tweets about contact tracing apps tell us about attitudes towards data sharing for public health? (Part 2)

In my last blog post I wrote about the importance of understanding attitudes towards data-sharing for public health and its link to my work with the LifeInfo Project. I also advocated that, by analysing Twitter conversations about using contact tracing apps to manage the spread of Covid-19, we could uncover some of people’s underlying thoughts, fears, and hopes about personal data sharing for better health.  

Analysis showed many to have taken to Twitter to share their thoughts, particularly sparked by the introduction of the Australian ‘COVIDSafe app’ and the Isle of Wight trial for the UK ‘NHS track and trace app’. In this blog, I will focus on the textual content of these tweets, using Natural Language Processing (NLP) and topic modelling, and discuss how this relates to opinions of data-sharing in the time of Covid-19. 

Hot Topics – what do tweets say?  

Continued tweet scraping means 18,170 tweets have now been collected on the topic of Covid-19 apps, between 24th April and 1st June 2020. As seen in my last blog, the most frequently used words in these tweets indicates that the conversation has been shaped around particular contexts. Nationalities and geographies are prominent – ‘Isle [of] Wight’, ‘Australian’, ‘UK’, ‘India’, ‘Icelanders’ – as are technological and state actors – ‘government’, ‘NHS’, ‘apple’, ‘google’. This introduces the idea that the wider conversation about Covid-19 apps is made up of smaller topic clusters about different places, specific apps, and associated news stories.  

Holly Clarke

Leeds Institute for Data Analytics

Holly Clarke is an Intern at Leeds Institute for Data Analytics, applying data science solutions to solve complex, real-world challenges. She is working for the LifeInfo project with Michelle Morris, researching attitudes towards novel lifestyle and health data linkages and how access to this information could improve public health. 

You can follow Holly on Twitter: @HollyEClarke

Word cloud showing online conversation around contact tracing apps

Topic Modelling is a technique often used to uncover latent topics within a set of documents (in this case tweets about Covid-19 apps) through statistical analysis of semantic similarities. The Latent Dirichlet Allocation (LDA) algorithm is built on the principle that each document is made up of small number of topics, and each topic is identifiable by its use of words.  

This algorithm was employed to uncover 12 topics within the Twitter conversation about Covid-19 apps, as displayed in Figure 1. A data-driven approach was used to identify the number of topics, taking the model with greatest probabilistic coherence across topics1. Table 1 has a full breakdown of topic interpretations with the words most associated with each topic displayed in Figure 1. 

Topics, however, need to be interpreted. Some consist of a finer granularity of related content, for example, Topic 9 focuses on symptom tracking apps, predominantly in the UK context but also in Kogi State, Nigeria. Other topics are modelled as separate but found to be lexically similar through hierarchical clustering. This is displayed by the dendrogram at the top of Figure 1, indicating how closely related the twelve topics are to each other2. For example, Topics 11 and 12 both focus on rights, and concerns about the security and privacy of coronavirus apps. Yet, the former takes a more technological perspective, using words like ‘cyber’, ‘tech’ and ‘digital’, while the latter uses more legal jargon – ‘legislation’, ‘laws’ ‘legal’.  

Topics within tweets about Covid-19 apps and the top 30 words associated with each topic according to their phi value.

Topic 1: google; apple; nhs; plan; rejects; uk; news; privacy; bbc; api; apps; tech; data; technology; model; tracing; centralised; government; approach; contact; decentralised; store; protect; british; governments; control; exposure; release; usage; insists.

Topic 2: government; care; workers; uk; phone; union; warns; private; access; data; install; installing; tracking; contact; contacts; personal; delete; list; staff; info; pay; smartphone; facebook; identify; messages; testing; intend; official; track; gmb.
Topic 3: isle; wight; nhs; source; code; nhsx; cummings; download; trial; uk; week; government; github; isleofwight; iow; android; trace; news; beta; matthancock; track; test; hancock; ios; matt; trialled; island; gov; iwnews; dominic.

Topic 4: privacy; india; surveillance; tracing; people; technology; helped; contact; icelanders; democracy; unlike; forcing; country; concerns; government; review; mandatory; countries; mit; tech; apps; launch; roll; setu; tracking; world; news; fears; france; aarogya.

Topic 5: download; government; auspol; trust; covidsafe; million; australia; downloads; australians; operational; tracing; downloaded; scottmorrisonmp; people; australian; morrison; capability; data; news; abc; federal; health; greghuntmp; govt; history; downloading; minister; lnp; police; covidapp.

Topic 6: health; tracker; city; chinese; permanent; plans; public; download; social; free; canada; track; china; support; information; medical; virus; pandemic; launched; resources; tracking; digital; distancing; access; time; dido; testing; officials; launches; residents.

Topic 7: phone; bluetooth; android; phones; people; contact; iphone; warn; nhs; location; security; uk; privacy; ios; battery; experts; background; person; running; creep; mission; positive; users; time; close; update; tracking; false; device; mobile.

Topic 8: download; live; safe; tracking; pm; stay; track; check; country; govt; india; people; choice; virus; news; avoid; report; home; questions; travel; days; update; freedom; millions; person; downloaded; forced; jobs; australia; indians 

Topic 9: people; test; symptoms; virus; lockdown; testing; day; weeks; 
report; daily; uk; data; reporting; users; govt; kogi; infection; positive;  risk; tests; research; tested; ago; reported; week; time; track; kings; pandemic; ministers.

Topic 10: people; download; government; downloading; phone; downloaded; facebook; social; twitter; virus; media; stop; track; govt; f**k; worried; text; phones; australia; time; australian; privacy; gov; telling; money; distancing; lol; f**king; sh*t; message. 

Topic 11: tracing; contact; privacy; nhs; security; uk; nhsx; data; symptoms; checking; experts; read; possibly; concerns; rights; expert; readies; technical; digital; post; tech; government; protection; cyber; design; apps; human; ncsc; world; qatar. 

Topic 12: data; nhs; google; privacy; information; tracing; personal; contact; access; government; apple; location; security; uk; drive; info; misuse; law; amazon; people; secret; laws; reveal; legislation; safeguards; legal; covidsafe; govt; post; north.
Figure 1: Topics within tweets about Covid-19 apps and the top 30 words associated with each topic according to their phi value (bottom), dendrogram indicating how lexically similar topics are to each other according to hierarchical clustering (top).  

Table 1 showing the interpretations of each topic, found through topic modellingdeveloped through examination of topic ‘top words and taking examples from original tweets 3 

1:  UK’s decision to develop a centralised app over the Google/Apple decentralised system. 
2: GNB union tells UK care workers not to use Coronavirus ‘Care Workforce app’ . 
3: Track and Trace app trialled on the Isle of White. Links with Dominic Cummings (Government Advisor).   
 4: Topics of privacy, force, democracy and surveillance relating to Indian ‘Aarogya Setu app’. ‘Icelanders’ appears, perhaps because Iceland was also an early adopter of a contact tracing app.   
5: Australians downloading COVIDsafe app. Political, with direct references to Scott Morrison (PM) and Greg Hunt (health secretary), ‘LNP’ (Liberal National Party), ‘police’ and ‘auspol’ (Australian politics hashtag). 
6: Several threads of news stories  within this cluster; Chinese city plans to transform Corona app into a permanent health tracker, people using a symptom tracker app in Canada, ‘Dido’ in relation to Dido Harding (ex-director of TalkTalk)’s involvement in NHS app. 
 7: Technological issues with contact tracing apps – Bluetooth, battery, background data, Android, Apple. Also, privacy, surveillance and ‘mission creep’ concerns with NHS app. 
 8: Broad topic with themes of safety – ‘stay’, ‘safe’, ‘home’ and mobility ‘freedom’, ‘travel’. 
9: Covid-19 Symptom Tracking app UK (Zoe) and research from King’s College with these data. Similar Kogi State (Nigeria) symptom self-assessment app. 
10: Parallels drawn between data collection by contact tracing apps and other social media platforms, negative words and expletives used, often in relation to irony, people rejecting apps due to privacy concerns but using Facebook etc. 
11: Personal rights and privacy/security concerns about coronavirus apps, from technological perspective ‘experts’ ‘technological’, ‘cyber’ 
12: Personal rights and privacy/security concerns about coronavirus apps, from a legal perspective – ‘misuse’, ‘access’, ‘laws’ ‘legislation’. 

Overall, topic modelling conveys several things about Covid-19 app attitudes. First, as hypothesised, conversations are strongly shaped around context, detailing prominent news stories and events, but this also influences how apps are talked about. Topic 4 includes references to issues of ‘privacy’, ‘surveillance’, ‘democracy’ and ‘force’ when talking about the Indian ‘aarogya setu app’ which is mandatory for government employees. Topics 1 and 7 have more of a technological and practical focus within the UK context, although still in relation to personal privacy, discussing centralised/decentralised apps, Bluetooth and battery issues. 

Second, although contact tracing apps are the predominant focus, topic modelling distinguishes other Covid-19 apps sharing different kinds of data.  Topic 1 focuses on symptom trackers and Topic 2 the ‘Care Workforce app’ to disseminate information.  

Third, most topics report events rather than represent attitudes or positions, although many do contain negative words such as ‘concern’, ‘warn’ or ‘worries’. Topic 10, however, stands out by including expletives – indicating anger is evident within this part of the conversation. 

All these topics make up a significant portion of the total tweets, ranging from 6.6% for Topic 2 to 10.6% for Topic 10, conveying that no single topic dominates.  

COVIDSafe vs Track&TraceActors Matter  

Alongside creating a dataset of tweets about Covid-19 apps I have also been collecting tweets about specific apps – the Australian ‘COVIDSafe app’, and the ‘NHS Covid-19 app’/‘Track&Trace app’, in the UK. My aim was to compare attitudes towards these apps and expose potential commonalities and differences between issues such as personal privacy, surveillance and data security, and link this to policies and practices. The overwhelming interpretation of this analysis, however, is that actors matter.  

Support or opposition to data sharing is greatly influenced by who we are giving access to these data and our trust in these actorsResearch consistently shows that, in the UK, we have the highest trust in the NHS, lower trust in central government and the lowest trust in private and technological companies.  In a blog post, Helen Kennedy, suggests that as contact tracing apps involve all three actors, the public will not know whether to trust them or not.  

Word clouds of most-frequent words included with tweets about the ‘Track and Trace app’ (UK) and the ‘COVIDSafe app’ (Australia).
Figure 2: Word clouds of most-frequent words included with tweets about the ‘Track and Trace app’ (UK) and the ‘COVIDSafe app’ (Australia).4 

The word ‘government’ appears prominently in both the COVIDSafe (Australia) and Track&Trace (UK) wordcloud as one of the most frequently used terms, showing the centrality of the state to conversations about contact tracing apps. Relatively high levels of trust in governments to use data appropriately means they may be in an advantageous position to convince people to share their personal data to help track and quell the spread of coronavirus. It is also possible that the ‘NHS’ branding of the UKs app could influence people to support and use it due to high levels of trust in this organisation. 

It would be remiss not to mention, however, the frequent references in the UK data to Dominic Cummings, the government chief advisor who during the time period of tweet collection has been scrutinised due to his journey to Durham during lockdown. Many frequently used words reference a rumour that his sister is involved with the contact tracing app – ‘Idox’, ‘sister’, ‘director’, ‘contract’, ‘Alice’.  Although this has been found to be untrue by the fact checking organisation FullFact, this association could undermine public trust and creates confusion about who has access to data collected from the app.  

Comparison of the number of tweets produced per day referencing the ‘NHS Covid-19 app’ and the ‘Track and Trace app’ (separate names for the UK app) between 30 April and 1 June 2020.
Figure 3: Comparison of the number of tweets produced per day referencing the ‘NHS Covid-19 app’ and the ‘Track and Trace app’ (separate names for the UK app) between 30 April and 1 June 2020. 

Word clouds of most-frequent words included with tweets about the ‘Track and Trace app’ (UK) between 29 April 2020 and 20 May 2020 (left) and 21 May to 1 June 2020 (right).
Figure 4: Word clouds of most-frequent words included with tweets about the ‘Track and Trace app’ (UK) between 29 April 2020 and 20 May 2020 (left) and 21 May to 1 June 2020 (right).4 

As is shown in Figure 3, tweets containing reference to a ‘NHS Covid-19 app’5 (the official naming of the UK’s app) are in line with tweets about a ‘Track and Trace app’ until mid-May. Past this point tweets about the NHS app are eclipsed by those about ‘Track and Trace’, showing the diminishment of the NHS branding at this point. At the same time, tweets about the ‘Track and Trace app’ contain fewer references to the ‘NHS’ and ‘government’ and greater references to ‘cummings’ (Figure 4).  This seems indicative of a shift in the public mind away from the app as a neutral, technological, health tool towards something more political.  

Comparatively, while references to Scott Morrison (PM) and Greg Hunt (Health Minister) are evident within the Australian dataset, the conversation seems to focus on the government as a unified actor. The UK have not announced a date for the track and trace app to be released nationally, but given downloads have fallen short of targets in Australia, it seems unlikely mass support will be mobilised here.  

In my next blog post I will be looking at what sentiment analysis on tweets can tell us about people’s attitudes to contact tracing apps,  follow up with any current developments, and round off this blog series with a discussion of how the Covid-19 pandemic might impact data-sharing practices and attitudes going forward and what this could mean from projects like LifeInfo.  

10 models were created with topics ranging from 1-101 in intervals of ten to estimate an optimal ‘topic window’ where topics were found to have the greatest probabilistic coherence. A further 20 models were then created within this window (11-31 topics) to select the model with the highest overall topic coherence.  

2  Closeness of topic found through hierarchical clustering using the Hellinger Distance for phi – (P(token|topic)) 

3Top words found by the highest phi values per topic, where phi is P(word|topic) 

common ‘stop words’ are excluded, for example ‘is’ or ‘and’, also the words directly related to the search terms for the app e.g. ‘covidsafe’ , ‘track’ or ‘app’. 

Inclusive of any reference to NHS ‘corona’ or ‘covid’ app.