In our research on Intelligent Systems and Data Science, we use semantic technologies to model and reason with large amounts of data, as well as machine learning and statistical techniques for mining knowledge from data in a variety of application domains. These include (but are not limited to) Smart Cities, Scholarly Data and Digital Humanities.
Furthermore, our interest lies on understanding what are the fundamental features of a “Science of Data”. Here, we focus on the way different types of data can be described to support their discovery, reuse and governance in complex processing infrastructures. In particular, in the context of developing intelligent metadata management applications, we have developed methods that can reason with semantic representations of policies and data flows to determine how licences and policies propagate upon complex data flows – see http://oro.open.ac.uk/52707/.
We are currently also working on developing new solutions for data cataloguing, aiming at improving the way such important assets are stored and managed as ‘libraries of data’. This work supports the data cataloguing infrastructure developed for the MK Data Hub, but it is of course, also applicable to other contexts.
To support Data Science research and development, the ISDS team manages the KMi Big Data Cluster, a processing infrastructure based on Apache Hadoop.
The ISDS team also develops and maintains the Open Knowledge Graph of The Open University and promotes an open approach to the dissemination and reuse of research outputs. Initiated in 2010, this was the first Open Knowledge Graph in UK academia leveraging Linked Data technologies in the education domain.