Top-k Exploration of Query Candidates for Efficient Keyword Search on Graph-Shaped (RDF) Data

Abstract

Keyword queries enjoy widespread usage as they represent an intuitive way of specifying information needs. Recently, answering keyword queries on graph-structured data has emerged as an important research topic. The prevalent approaches build on dedicated indexing techniques as well as search algorithms aiming at finding substructures that connect the data elements matching the keywords. In this paper, we introduce a novel keyword search paradigm for graph-structured data, focusing in particular on the RDF data model. Instead of computing answers directly as in previous approaches, we first compute queries from the keywords, allowing the user to choose the appropriate query, and finally, process the query using the underlying database engine. Thereby, the full range of database optimization techniques can be leveraged for query processing. For the computation of queries, we propose a novel algorithm for the exploration of top-k matching subgraphs. While related techniques search the best answer trees, our algorithm is guaranteed to compute all k subgraphs with lowest costs, including cyclic graphs. By performing exploration only on a summary data structure derived from the data graph, we achieve promising performance improvements compared to other approaches.

KOIOS: Utilizing Semantic Search for Easy-Access and Visualization of Structured Environmental Data

“…Therefore, we rely on a top-k procedure to generate candidate interpretations and obtain the possible results that best match the user information need. We build on our previous work [17] on translating keyword queries into structured queries based on a graph-exploration technique. For this purpose, we consider available knowledge bases and data as data graphs as defined previously.…”

“…It has been shown that keyword search is most efficient when the exploration for possible interpretations is performed on an augmented schema graph, instead of using the entire data graph (c.f. [17]). The schema graph can be trivially obtained from the class and property definitions in the data, or might be pre-given as an ontology.…”

“…The computation of structured queries as interpretations of the user keywords involves three tasks: 1) detecting the keyword elements as resources in the graph containing at least one of the query keywords, 2) creation of an augmented schema graph, 3) graph exploration and structured query generation and 4) topk query processing to compute the best queries. Specific concepts and algorithms for these tasks have been introduced in [17].…”

“…By utilizing lightweight semantics of the underlying RDF data, keyword search can help to circumvent the complexity of structured query languages, and hide the underlying data representation. Without knowledge of the query syntax and data schema, even the non-technical end-users can obtain complex structured results, including complex answers generated from RDF resources [17,19]. To this end, semantic search in that sense can provide the means to simplify the search on environmental data, allowing the users to access rich information with less effort, and lower the access barriers resulting from the complex interfaces of current systems.…”

Entity Reference Resolution via Spreading Activation on RDF-Graphs

“…The recent approach BLINKS [12], based on [13], selects clusters according to the smallest cardinality. Thanh et al [23] presented a search algorithm for RDF-Graphs based on one iterator per surrogate set. Recently, Kasneci et al [12] proposed an approximation algorithm for Steiner trees that allows for a fast retrieval of relationship graphs based on minimal weights.…”