An Agglomerative Query Model for Discovery in Linked Data: Semantics and Approach

Data on the Web is increasingly being used for discovery and exploratory tasks. Unlike traditional fact-finding tasks that require only the typical single-query and response paradigm, these tasks involve a multistage search process in which bits of information are accumulated over a series of related queries. The ability and effectiveness of users to connect the dots between these pieces of information are crucial to enable discovery. In this paper, we introduce the notion of agglomerative querying for supporting “search processes” and present its motivation, challenges and formalization. We focus on a specific class of agglomerative querying called association agglomerative querying which is very natural for linked data models such as RDF. We present a preliminary implementation approach for processing such queries and discuss its relationship with SPARQL query processing. Finally, we present empirical results for proving the effectiveness of our approach on the DBLP dataset and ...

• Type: Conference
• Year: 2010
• Where: WEBDB
• Authors: Sidan Gao, Haizhou Fu, Kemafor Anyanwu
• Keywords: Database, Exploratory Tasks, Query Processing, Traditional Fact-finding Tasks, WEBDB 2010