Publishing information as Linked Data enables Government Agencies to provide context around data, so that information can be understood, searched and shared more easily. Technically speaking, this approach addresses the requirement to deliver semantic interoperability besides basic technical interoperability. It allows users to seamlessly search and combine data from diverse sources and to derive new knowledge by means of reasoning, which is becoming increasingly important in the context of Machine Learning. In this session, we will explain how Ordnance Survey Ireland (OSi) extended their existing IT environment to publish datasets as Linked Data. OSi, the Central Statistics Office and the ADAPTcentre at Trinity College Dublin collaborated on the project with the goal of laying the foundations of a semantic architecture. In our talk, we will go over the fundamentals of Linked Data and the benefits and challenges associated with it. Using administrative boundary data as an example, we will describe how mostly existing ontologies were used to structure the data, how the datasets were mapped to these structures and how the data can be queried and blended with other information. We will look at the specific requirements of geospatial data and how they are taken into account in the different components of a Linked Data platform, using standards developed by the Open Geospatial Consortium. A summary of the lessons learnt and an outlook on future directions are included to round off the session.
Start date: Thursday, 10 October 2019 | End date: Friday, 11 October 2019
All content © Delegate Live Limited. All rights reserved.