A Novel Metric for Linked Data Analysis: A Case Study on Predicting the US Presidential Election Results!

Conference Paper
R. Meymandpour
2012, November
Published in: 
24th Research Conversazione
Published in: 
The University of Sydney
According to the Linking Open Data (LOD) principles, datasets from various domains are semantically described and interlinked. This publicly available semantic data sources opens up many opportunities for further knowledge extraction, analysis and interpretation. In this research, we address the problem of automatic consumption and analysis of Linked Data by proposing a novel metric to measure the informativeness of the Web of Data and its resources. Informativeness of an item is the amount of valuable information conveyed by the item. The presented metric has several potentials to be employed in innovative semantic applications such as faceted browsing, recommendation provision and entity ranking.