News round-up: May 2018
VisitScotland: Unlocking the Power of Data
Over the past month we’ve been working with VisitScotland to advise them on the most efficient and cost-effective ways to work with their data.
VisitScotland works to maximise the visitor economy by creating strategies and insights that support economic growth and investment. To do this they pull together a very diverse range of information sources – including national and international tourism statistics, survey data and research projects. While they have a wealth of valuable information, it’s often time-consuming to locate, format and pull together. Many organisations face similar challenges with big data, and are looking for smarter ways to collect, process and publish an ever-increasing amount of information. We’ve been busy researching VisitScotland’s data landscape, workflows and stakeholders in order to provide recommendations and advice.
If you’d like to turn your Big Data challenge into an opportunity, Wallscope can help! Get in touch to find out more about our consultancy and advisory services.
Statistical modelling with SynthLab
We’ve been continuing our collaborative R&D project with Edinburgh University Business School, funded by The Data Lab. The product we’re evolving – SynthLab – uses a statistical computing platform (called ‘R’) to interpret large amounts of data and analyse the results in various ways.
The platform works by clustering the data into groups that have similar properties – in this case we are clustering GP practices. Using data models based on these groups, it is possible to create visualisations and predictive models. These can be used to inform decision making and to feed business intelligence tools - for example, to predict where demand for a particular drug will rise or fall over time.
This is an extension of work we did with the Information Services Division (ISD) of NHS National Services Scotland following our time with them on the CivTech pilot, developing dashboards that join data on prescriptions, drugs and the NHS workforce.
Due to the sensitive nature of patient information, we make use of ‘synthetic data’ – realistic data sets that can be released without the risk of identifying individuals. Using our platform, completely anonymous yet realistic data could be released to drive innovation in the NHS in Scotland.
Wallscope’s Data Foundry is a document indexing service that reads electronic documents, such as Word, PDF, CSV, XLS and email. It uses Natural Language Processing to find entities within a document, and then links it to related people, places, organisations and concepts.
The current development focus is on scalability so we can index documents faster and with greater accuracy.
Our Lead Developer Antero Duarte says: ‘This project is very interesting because it poses challenges as the size of data grows, essentially stress-testing our whole software stack and making us push it to the next level. It’s even more interesting to realise how this could become an essential tool for a company to organise and unlock the potential of their unstructured data in ways we have not seen before.’
An obvious use case would be to link legacy or siloed data, though connecting to live data streams is also possible. A combination of the two would help to create a truly dynamic data resource. Keep your eyes open for a demo release in the coming weeks!