Now for Something Data Science Ep.2

This is a hybrid event - join in person or online with MS Teams.

The 2nd instalment of our data science mini-series showcasing current research projects carried out by the Data Scientist Development Programme at Leeds Institute for Data Analytics (LIDA).

This week’s features 10-minute lightning talks from three data scientists on online food shopping, illegal drug markets and health.

Each talk will last 10mins + 10mins Q&A.

Agenda

Talk 1 - Finding Healthy Online: What the Eyes Reveal

(Gauri Venkatachalapathi)

This data science project, analyses and visualises eye-tracking data collected from ten Ocado shoppers to understand customer online purchase choices, supermarket website use, and browsing patterns. Eye tracking (fixations and saccades) data was analysed alongside survey and purchase information on the dashboard to visually explore how consumers navigate and engage with the supermarket website elements and information. Data sources were consolidated and curated (Excel and R) to graphically visualise insights with a Quarto Dashboard. Findings will support Ocado’s wider website design to encourage healthier and more sustainable food choices.

Talk 2 - Psychiatric History and Routinely Assessed Outcomes of Hospitalisation (PHARAOH)

(Precious-Gift Alele)

People with both mental and physical health conditions (Mental-Physical Multimorbidity, MPM), face significant challenges in hospital care, but little is known about how their health outcomes differ compared to people with only a physical health condition.

Using the Leeds Data Model (LDM) we tracked patient journeys from primary care through hospital admission, back to primary care. We were able to compare outcomes in MPM versus non-MPM patients over 5 years. Unadjusted results showed substantial differences. The success of our approach will permit more detailed analyses of individuals and diagnoses to develop ways to reduce this important health inequality.

Talk 3 - Analysing Property-based Illegal Drug Market Location Choices

(Angeliki Fragkeskou)

This talk investigates how area-level characteristics influence the siting of property-based illegal drug markets in the area of Leeds. Offenders increasingly repurpose residential and commercial properties as hidden hubs for drug activities, often exploiting socio-economic vulnerabilities such as deprivation, high population density, and poor access to services. By combining demographic and environmental data at the Census Output Area level, the study identifies the factors that significantly increase the likelihood of an area being targeted for illegal activities. The findings aim to equip local authorities and community safety teams with evidence-based insights to design targeted interventions that prevent criminal exploitation and reduce harm to communities.

More about LIDA

Sign up to hear about LIDA events

Find out more about the Data Scientist Development Programme