Data Analytics for Future Cities

Lead Research Organisation: University of Strathclyde
Department Name: Mathematics and Statistics

Abstract

The "Internet of Things" is a phrase describing the set of technologies, systems and methodologies that underpin the spread of internet-enabled applications. Ultimately, the Internet of Things will involve physical objects seamlessly integrating into the information network for social and economic benefit.

At the heart of the Internet of Things is data---digital records of human, technological and natural interactions. The data streams are large-scale, varied and rapidly changing. In addition to the important, but conceptually simple, tasks of gathering, storing and sharing this data, there is a pressing need to develop powerful and efficient computational algorithms that can extract insights and make useful predictions.

This proposal will add to the "cleverness" that is needed to exploit fully the data deluge. Technology evolves rapidly. The look, feel and functionality of smartphones, tablets, notebooks and desktop PCs have changed dramatically in recent years, and there is a range of new technologies such as wearable devices, smart glasses and implantable sensors. Many of these will add to the technological revolution, before themselves being superseded. However, the challenge of analysing and exploiting the vast realms of data produced by the Internet of Things is universal, and the mathematical concepts and resulting algorithms that underpin these new technologies are fundamental and have very long-term value.

My proposal will develop new analytical concepts that lead to computational algorithms. The results will be aimed directly at the application of Future Cities research---improving the social, environmental and economic aspects of city living. The University of Strathclyde is home to a City Observatory that collects a huge amount of data from the city of Glasgow, including air quality sensors, traffic information, energy usage and measurements of many aspects of human behaviour, such as on-line activity, social media interactions, CCTV data and retail footfall counts. Indeed, Glasgow received £24M of government funding to become the UK's pilot city for this type of digitally-driven enhancement.

The research project therefore focusses on new concepts and algorithms that can help us to understand the very large, fast-moving streams of data describing the interactions between the components that make up the Internet of Things: people, devices and sensors. Just as Google began with a clever mathematical algorithm, PageRank, that was able to bring order to the world wide web, we aim to develop new algorithms that can summarize these vast quantities of data. The work will take place alongside stakeholders in the Future Cities arena: fellow-researchers in social science; SMEs who deliver data analytics solutions to clients in advertising, finance, entertainment, publishing; external partners in hi-tech industry who deliver larger-scale IT solutions; local and national government employees who serve the community. Through a range of knowledge exchange and outreach events, these stakeholders will have the opportunity to critically evaluate and feedback on the results and rapidly deploy the new ideas.

Some illustrative applications that the research will address are:

characterising the social media traits of different user bases, such as drivers/cyclists/pedestrians, to predict the best way to target messages at each group,

stratifying the population of city users according to their portfolio of work/leisure/shopping community memberships in order to maximise the usage of energy/space resources in the city,

predicting crowd levels and crowd behaviour at forthcoming public events,

monitoring the public perception of an ongoing campaign, such as a cycle-to-work-scheme,

monitoring and reacting in real time to the public response in relation to a planned disruption, such as a political march, or an unpredictable event, such as a traffic incident.

Planned Impact

A recent Deloitte report, "Measuring the Economic Benefits of Mathematical Science Research," estimated that analytics of big data will contribute £40 Billion to the UK economy in 2017 and create 58,000 new jobs over the next 6 years. In Policy Exchange, the UK's leading think tank, a report by David Willets lists The Big Data Revolution and Energy-Efficient Computing as the first of the eight prioritized technologies.

This proposal will deliver impact in these burgeoning fields by contributing innovative, general purpose, industry-tested, tools for data analytics that allow more efficient and intelligent use of the data streams concerning human behaviour and the natural and technological environment, including social media, traffic, public transport, energy usage, crime, air quality, retail footfall.

The project is embedded within a Future Cities framework. Impact will be monitored as the project proceeds via direct interaction with external partners and their clients/customers. Immediate beneficiaries will include

SMEs who develop bespoke data analytics applications for clients in advertising, entertainment, energy, local and national government, such as Bloom and CountingLab, as highlighted in their Letters of Support, who will gain access to more insightful data analytics tools that can be customized to their needs,

larger hi-tech companies developing complex IT solutions across multiple platforms; as outlined in the Letters of Support from Siemens and Capita, who will be exposed to cutting edge research ideas in a general purpose, widely applicable setting,

clients of these companies, including retailers, transport companies, energy and other utilities, charities and governments, who will benefit from the extra efficiencies and insights through improved products and services; Scotrail has provided a Letter of Support and Capita's letter notes their strong connections with IBM, Cisco and Dell,

policy makers, who will be better placed to mke informed decisions about their cities; Glasgow City Council recently won a £24M 'Future City' award from TSB and has a close working relationship with Strathclyde's Institute for Future Cities,

user-communities living, working or spending leisure time in a city environment, who will experience better services and who will be encouraged to make more positive lifestyle choices.


The aim of creating impact has informed each theme in the proposed project, as elaborated in the Case for Support, and key examples where the beneficiaries will gain include

allowing service providers to classify their user bases and discover influential or receptive subgroups, so that important messages can be targeted at key sections,

stratifying these user-bases in real time, so that providers can predict changes in demand over space and time,

gauging the current opinion of a community in order to predict response to a planned action or event, and to monitor public perception of a on-going campaign,

reacting effectively to real-time crises or news events through increased understanding of human behaviour during a social media spike, and exploiting the resulting opportunities to intervene positively with customized messaging,

predicting fluctuations in demand from city-centre shoppers for resources such as energy, transport and wi-fi, in order to maximise efficiency and avoid failures,

understanding which collection of data streams are most informative about levels of crime or anti-social behaviour, in order to suggest how resources should be targeted most efficiently.


Further, by training a PDRA, the project will also deliver a skilled, outward facing, future-leader with experience of interacting with external partners. This individual will be well placed to act as an agent-for-change in shaping the UK demand for Data Analytics in way that strengthens UK competitiveness.

Publications

10 25 50

publication icon
Anderson D (2019) On Constrained Langevin Equations and (Bio)Chemical Reaction Networks in Multiscale Modeling & Simulation

publication icon
Arrigo F (2019) Non-backtracking PageRank in Journal of Scientific Computing

publication icon
Arrigo F (2018) Non-backtracking walk centrality for directed networks in Journal of Complex Networks

publication icon
Arrigo F (2018) On the exponential generating function for non-backtracking walks in Linear Algebra and its Applications

publication icon
Arrigo F (2017) Sparse matrix computations for dynamic network centrality. in Applied network science

publication icon
Arrigo F (2019) Non-Backtracking Alternating Walks in SIAM Journal on Applied Mathematics

publication icon
Fenu C (2017) Block Matrix Formulations for Evolving Networks in SIAM Journal on Matrix Analysis and Applications

publication icon
Grindrod P (2018) High Modularity Creates Scaling Laws. in Scientific reports

publication icon
GRINDROD P (2016) Inverse network sampling to explore online brand allegiance in European Journal of Applied Mathematics

Related Projects

Project Reference Relationship Related To Start End Award Value
EP/M00158X/1 01/01/2015 31/03/2019 £643,306
EP/M00158X/2 Transfer EP/M00158X/1 01/04/2019 30/06/2020 £67,241
 
Description A range of new computational tools that can be used to extract useful information from lareg scale dat asets arising from human digital behavious.
Exploitation Route Use of software provided with academic publications. Implementation of the models and algorithms that have been developed.
Sectors Digital/Communication/Information Technologies (including Software),Energy,Retail,Security and Diplomacy,Transport,Other

URL http://personal.strath.ac.uk/d.j.higham/webcv.pdf
 
Description With Peter Grindrod (Oxford) I continue to collaborate with a digital marketing in Leeds (Bloom). We recently produced a (public domain) research article "Inverse Network Sampling to Explore On-line Brand Allegiance", to appear in European Journal of Applied Maths, and a chapter of an edited book on "Success Stories in UK Applied Maths", see publications. I also took part in a "Meet the Scientist" session at Glasgow Science Centre in 2015 and accepted an invitation to join a discussion panel on Maths/Future Cities at the Edinburgh Science Festival in 2016. I guest edited a special issue of the Royal Society Open Science journal on City Analytics, wich had a launch-day event at the Alan Turing Institute. I also co-organised a one-day workshop at the Royal Society of Edinburgh on the same theme. In 2018 I coorganised a one-week research/industry event at the International Centre for Mathematical Sciences on Mathematics and Measurement, which involved 50% attendance from national and international government metrology labs. In 2018 I also coorganised workshop son Data Science & Social Media, and on Data Science and Crime.
First Year Of Impact 2017
Sector Digital/Communication/Information Technologies (including Software),Other
Impact Types Societal,Economic

 
Description EPSRC Mathematical Sciences Programme Grant
Amount £2,964,066 (GBP)
Funding ID EP/P020720/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Academic/University
Country United Kingdom
Start 06/2017 
End 05/2022
 
Title MATLAB Software for Deep Learning 
Description Scientific Computing software to accompany an expository article on the mathematics of deep learning 
Type Of Material Improvements to research infrastructure 
Year Produced 2018 
Provided To Others? Yes  
Impact Accompanying expository article has been used in several universities as the basis for classes, tutorials and reading groups on deep learning. 
URL http://personal.strath.ac.uk/d.j.higham/algfiles.html
 
Title New Algortihm for Core-Periphery Detection 
Description Implemntation of a newly published algorithm that discovers importsnt substructures in large scal enetworks 
Type Of Technology New/Improved Technique/Technology 
Year Produced 2019 
Impact Paper has not yet appeared, so current impact is limited to closely-related research groups. 
URL https://github.com/ftudisco
 
Description Workshop on Data Science and Crime 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact One day workshop involving police data scientists from West Midlands, Police Scotland and ondon Met, and academics with relevant skills. Knowledge exchange and highlighting of current challenges were the key aims.
Year(s) Of Engagement Activity 2018
URL https://iconicmath.org/ds-crime-workshop/
 
Description Workshop on Data Science and Soaicl Media 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Industry/Business
Results and Impact An afternoon Workshop on Data Science and Social Media, held as part of a wekk ling Engage with Strathclyde series of events.
About 30 people attended from social media, policymaking, economics, to share best practice and learn about new developments at the intersection of Data Science and Social media.
Year(s) Of Engagement Activity 2018
URL https://www.engage.strath.ac.uk/