Skip to main content

Identifying Knowledge Anchors in Data Graphs

Venue: Birkbeck Main Building, Malet Street

No booking required

The recent growth of the Web of Data has brought to the fore the need to develop intelligent means to support user exploration through big data graphs. It has been acknowledged that, to be effective, approaches for data graph exploration should take into account the knowledge utility of exploration paths – how useful the trajectories in a data graph are for expanding users’ knowledge. Motivated by an earlier controlled user study investigating nudging strategies for exploration, which has suggested that paths which start with familiar and highly inclusive entities and bring something new are likely to have good knowledge utility, we propose here an approach to identify knowledge anchors in a data graph. We call such anchors basic level entities in a data graph, following an analogy with basic level objects in domain taxonomies that underpin our approach. Several metrics for extracting basic level entities in a data graph, and the corresponding algorithms, have been developed. The performance of the metrics is examined using benchmarking sets obtained from an experimental study involving free naming tasks by humans. Based on quantitative and qualitative analysis of the individual metrics, a hybridization approach is proposed.


Marwan Al-Tawil is a third year postgraduate researcher in the Artificial Intelligence Group in the Computing Department at the University of Leeds. His research interests lie in the field of graph databases, particularly in developing computational methods and algorithms to support users exploration over big data graphs.

Contact name: