A Path-Oriented RDF Index for Keyword Search Query Processing

Abstract

"Most of the recent approaches to keyword search employ graph structured representation of data. Answers to queries are generally sub-structures of the graph, containing one or more keywords. While finding the nodes matching keywords is relatively easy, determining the connections between such nodes is a complex problem requiring on-the-fly time consuming graph exploration. Current techniques suffer from poorly performing worst case scenario or from indexing schemes that provide little support to the discovery of connections between nodes. In this paper, we present an indexing scheme for RDF that exposes the structural characteristics of the graph, its paths and the information on the reachability of nodes. This knowledge is exploited to expedite the retrieval of the sub-structures representing the query results. In addition, the index is organized to facilitate maintenance operations as the dataset evolves. Experimental results demonstrates the feasibility of our index that significantly improves the query execution performance.

Path index based keywords to SPARQL query transformation for semantic data federations

“…But Cappellari et al [7] mentions that keywords can mostly reside on sinks. Sink is a node in RDF graph which does not have any outgoing edges from it.…”

“…Many approaches [6], [8], [9] try to find suitable sub-graphs by graph traversing at real time, which is highly time consuming. A path store [7] is utilized in this research for sub-graph identification because it has totally eliminated real time graph traversal.…”

“…Path Index assumes that user queries targets only sinks [7]. Therefore, only data about sinks are stored in Node table.…”

“…BLINKS [5] believes that keywords can only occur in vertices of the graph so index only vertices. Path Index [7] only store sink nodes in their index arguing that keywords only reside in sink nodes. When storing details, most approaches index only the label of the graph element.…”

“…Tran et al [10] suggest a graph based model to sub-graph identification. Cappellari et al [7] suggest a path index based approach, which stores edge sequences from source nodes to sink nodes of RDF graphs. As they have totally eliminated graph traversing at real time, efficiency of this approach is really high.…”