A Novel Method of Keyword Query for RDF Data Based on Bipartite Graph
Zhiyun Zheng, Yang Ding, Zhentao Wang, Zhenfei Wang · 2016
As huge amounts of the semantic Web data have sprung up, RDF data query becomes an important research topic. RDF data based on graph structure can keep correlation information and semantic information, so more and more keywords query methods model RDF data as RDF graph. But current query techniques on graph suffer from several drawbacks, like low precision, high query response time, high parallel implementation cost, and so forth. To address these problems, a novel method of keyword query for RDF data based on bipartite graph is proposed. Specifically, we first construct RDF data as bipartite graph with node labels in which all text information is encapsulated to support relationship query. And then we design a keyword expansion query algorithm which includes keywords expansion, keyword matching, and the construction of query result subgraphs. Moreover, the keyword expansion technology effectively solves the problem of delivering the same object description words and also improves the query precision. Finally, in experiments using large real-world dataset, our solution outperformed the state-of-the-art in terms of precision and query response time.