Discovery Hub: exploratory search on remote linked data sources
Exploratory search systems are built specifically to help the users during cognitive consuming search tasks like learning or topic investigation. The richness of the linked data datasets, their formal semantics and their connected nature can significantly help in this context. In this paper, we present a method to perform exploratory search on remote LOD sources. It uses a linked-data based recommender that identifies resources which are strongly related to the users topic of interest. This recommendation step aims to ease the exploratory search task by focusing the users attention on the results that convey a lot of knowledge. To perform it on remote data sources we propose a semantic spreading activation algorithm coupled with a graph sampling technique. This paper also details the analysis of the algorithm behavior over DBpedia and describes an implementation: the Discovery Hub application. Finally the results of a user evaluation are presented.