Elsevier

Journal of Web Semantics

Volume 7, Issue 3, September 2009, Pages 166-176
Journal of Web Semantics

From keywords to semantic queries—Incremental query construction on the semantic web

https://doi.org/10.1016/j.websem.2009.07.005Get rights and content

Abstract

Constructing semantic queries is a demanding task for human users, as it requires mastering a query language as well as the schema which has been used for storing the data. In this paper, we describe QUICK, a novel system for helping users to construct semantic queries in a given domain. QUICK combines the convenience of keyword search with the expressivity of semantic queries. Users start with a keyword query and then are guided through a process of incremental refinement steps to specify the query intention. We describe the overall design of QUICK, present the core algorithms to enable efficient query construction, and finally demonstrate the effectiveness of our system through an experimental study.

Introduction

With the advance of the semantic web, increasing amounts of data are available in a structured and machine understandable form. This opens opportunities for users to employ semantic queries instead of simple keyword based ones, to accurately express the information need. However, constructing semantic queries is a demanding task for human users [10]. To compose a valid semantic query, a user has to (1) master a query language (e.g., SPARQL) and (2) acquire sufficient knowledge about the ontology or the schema of the data source. While there are systems which support this task with visual tools [20], [25] or natural language interfaces [4], [12], [13], [17], the process of query construction can still be complex and time consuming. According to [23], users prefer keyword search, and struggle with the construction of semantic queries although being supported with a natural language interface.

Several keyword search approaches have already been proposed to ease information seeking on semantic data [15], [31], [33] or databases [1], [30]. However, keyword queries lack the expressivity to precisely describe the user’s intent. As a result, ranking can at best put query intentions of the majority on top, making it impossible to take the intentions of all users into consideration.

In this paper, we introduce QUICK,1 a novel system for querying semantic data. QUICK internally works on pre-defined domain-specific ontologies. A user starts by entering a keyword query, QUICK then guides the user through an incremental construction process, which quickly leads to the desired semantic query. Users are assumed to have basic domain knowledge, but don’t need specific details of the ontology, or proficiency in a query language. In that way, QUICK combines the convenience of keyword search with the expressivity of semantic queries.

The paper presents a detailed realization of the QUICK system, including the following contributions: (1) we defined a framework for incrementally constructing semantic queries from keywords; (2) we devised algorithms to generate near-optimal query construction guides, which enable users to quickly construct semantic queries; (3) to support the QUICK system, we designed a scheme for optimizing the execution of full-text queries on RDF data; (4) we conducted experiments to evaluate the effectiveness of QUICK and the efficiency of the proposed algorithms.

The rest of the paper is organized as follows. Section 2 provides an overview on how to use QUICK. In Section 3 we present our framework for incremental query construction. Section 4 presents algorithms for generating near-optimal query guides. Section 5 introduces optimization techniques to improve query execution performance. In Section 6, we present the results of our experimental evaluation. Section 7 reviews the related work. We close with conclusions in Section 8.

Section snippets

QUICK overview

As illustrated in Fig. 1, the interface of QUICK consists of three parts, a search field (on the top), the construction pane showing query construction options (on the left), and the query pane showing semantic queries (on the right).

Suppose a user looks for a movie set in London and directed by Egdar Wright.2 The user starts by entering a keyword query, for instance ‘wright london’. Of course, these

Query construction framework

In this section, we introduce the query construction framework of QUICK. We describe our model for transforming keyword queries to semantic queries using an incremental refinement process.

Query guide generation

For a given keyword query, multiple possible query guides exist. While every guide allows the user to obtain the wanted semantic query, they differ significantly in effectiveness as pointed out in Section 3.3. It is thus essential to find a guide that imposes as little effort on the user as possible, i.e., a minimum query guide. Query construction graphs have several helpful properties for constructing query guides:

Lemma 10 Query construction graph properties

  • (i)

    Given a node in a query construction graph, the complete sub-graph with this

Query evaluation

When the user finally selects a query that reflects the actual intention, it will be converted to a SPARQL query and evaluated against an RDF store to retrieve the results. The conversion process is straight forward: For each concept node in the query or edge between nodes, a triple pattern expression of SPARQL is generated. In the first case, it specifies the node type, in the second case it specifies the relation between the nodes. Finally, for each search term, a filter expression is added.

Experimental evaluation

We implemented the QUICK system using Java. The implementation uses Sesame2 [5] as RDF Store and the inverted index provided by LuceneSail [18] to facilitate semantic query generation. Parts of the described query optimization approaches have been integrated to Sesame2 version 2.2. We have used this implementation to conduct a set of experiments to evaluate the effectiveness and efficiency of the QUICK system and present our results in this section.

Related work

In recent years, a number of user interfaces have been proposed to facilitate construction of semantic queries. These interfaces can be mainly classified into visual graphic interfaces [20], [25] and natural language interfaces [13], [4], [17]. Natural language interfaces are potentially more convenient for end users [12], as they require little prior knowledge of the data schema or a query language. However, the state-of-the-art natural language interfaces still require users to use a

Conclusion

In this paper, we introduced QUICK, a system for guiding users in constructing semantic queries from keywords. QUICK allows users to query semantic data without any prior knowledge of its ontology. A user starts with an arbitrary keyword query and incrementally transforms it into the intended semantic query. In this way, QUICK integrates the ease of use of keyword search with the expressiveness of semantic queries.

The presented algorithms optimize this process such that the user can construct

References (33)

  • V. Lopez et al.

    Aqualog: an ontology-driven question answering system for organizational semantic intranets

    Journal of Web Semantics

    (2007)
  • S. Agrawal et al.

    DBXplorer: a system for keyword-based search over relational databases

  • F.L. Clement et al.

    Effective keyword search in relational databases

  • H. Bast et al.

    The CompleteSearch engine: interactive, efficient, and towards IR & DB integration

  • A. Bernstein et al.

    Gino—a guided input natural language ontology editor

  • J. Broekstra et al.

    Sesame: a generic architecture for storing and querying RDF and RDF Schema

  • V. Broughton et al.

    Classification schemes revisited: applications to web indexing and searching

    Journal of Internet Cataloging

    (2000)
  • H. Haller

    QuiKey—the smart semantic commandline (a concept)

  • A. Harth et al.

    Optimized index structures for querying RDF from the Web

  • A. Harth, J. Umbrich, A. Hogan, S. Decker, YARS2: a federated repository for querying graph structured data from the...
  • H.V. Jagadish, A. Chapman, A. Elkiss, M. Jayapandian, Y. Li, A. Nandi, C. Yu, Making database systems usable, in:...
  • R.M. Karp

    Reducibility among combinatorial problems

  • E. Kaufmann, A. Bernstein, How useful are natural language interfaces to the semantic web for casual end-users, in:...
  • E. Kaufmann, A. Bernstein, R. Zumstein, Querix: a natural language interface to query ontologies based on clarification...
  • B. Kimelfeld, Y. Sagiv, Finding and approximating top-k answers in keyword proximity search, in: PODS,...
  • Y. Lei, V.S. Uren, E. Motta, Semsearch: a search engine for the semantic web, in: EKAW,...
  • Cited by (128)

    View all citing articles on Scopus
    View full text