Linked Open Data provides a means of unified access to large and complex interconnected data sets that concern themselves about arbitrary facts and events. This unified access in turn allows for the consumption of this data for modelling cultural heritage sites, historical events and complex problem solving. The cost comes in that traditional approaches to information retrieval, discovery and relevance are still strongly wedded to tabular or plain text data. How to consume what amounts to graphs of graphs with a variety of types and serendipitous annotations? Current applications rest on an assumption that locating the appropriate dataset is a straightforward function, which is becoming a research problem in itself.
I will present ongoing work on simulating the terrain and events of a Great War battle using data from multiple Linked Open Data projects. The ability to change back and forth between the statistical estimation of parameters and its inference from "big data" will be reviewed.
Dr Warren is a research scientist at the Big Data Institute and a research associate with the Canadian Writing Research Collaboratory focusing on Linked Open Data, ontologies, social network analysis, knowledge discovery in databases and ontologies. An engineer by training, he had previously held research posts in industry, government and academia both in North America and in Europe.
Last modified: Wednesday, 08-Jul-2015 14:46:16 NZST
This page is maintained by the seminar list administrator.