A Visual Interface for Building SPARQL Queries in Konduit.
- SourceAvailable from: deri.ie[show abstract] [hide abstract]
ABSTRACT: In this paper, we discuss the architecture and implementation of the Semantic Web Search Engine (SWSE). Following traditional search engine architecture, SWSE consists of crawling, data enhancing, indexing and a user interface for search, browsing and retrieval of information; unlike traditional search engines, SWSE operates over RDF Web data – loosely also known as Linked Data – which implies unique challenges for the system design, architecture, algorithms, implementation and user interface. In particular, many challenges exist in adopting Semantic Web technologies for Web data: the unique challenges of the Web – in terms of scale, unreliability, inconsistency and noise – are largely overlooked by the current Semantic Web standards. Herein, we describe the current SWSE system, initially detailing the architecture and later elaborating upon the function, design, implementation and performance of each individual component. In so doing, we also give an insight into how current Semantic Web standards can be tailored, in a best-effort manner, for use on Web data. Throughout, we offer evaluation and complementary argumentation to support our design choices, and also offer discussion on future directions and open research questions. Later, we also provide candid discussion relating to the difficulties currently faced in bringing such a search engine into the mainstream, and lessons learnt from roughly six years working on the Semantic Web Search Engine project.J. Web Sem. 01/2011; 9:365-401.
- [show abstract] [hide abstract]
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.Web Semantics: Science, Services and Agents on the World Wide Web. 09/2009;
Conference Proceeding: A Structured Semantic Query Interface for Reasoning-Based Search and Retrieval.[show abstract] [hide abstract]
ABSTRACT: Information and knowledge retrieval are today some of the main assets of the Semantic Web. However, a notable immaturity still exists, as to what tools, methods and standards may be used to effectively achieve these goals. No matter what approach is actually followed, querying Semantic Web information often requires deep knowledge of the ontological syntax, the querying protocol and the knowledge base structure as well as a careful elaboration of the query itself, in order to extract the desired results. In this paper, we propose a structured semantic query interface that helps to construct and submit entailment-based queries in an intuitive way. It is designed so as to capture the meaning of the intended user query, regardless of the formalism actually being used, and to transparently formulate one in reasoner-compatible format. This interface has been deployed on top of the semantic search prototype of the DSpace digital repository system.The Semantic Web: Research and Applications - 8th Extended Semantic Web Conference, ESWC 2011, Heraklion, Crete, Greece, May 29-June 2, 2011, Proceedings, Part I; 01/2011
A Visual Interface for Building SPARQL Queries in Konduit
National University of Ireland,
Digital Entreprise Research
This short demo description presents an extension to the
Konduit tool for visual programming for the semantic desk-
top. Previously, Konduit required users to define filter com-
ponents by manually specifying a SPARQL CONSTRUCT
query. With the current work presented in this demo, we are
exploring ways of building such queries visually, and aiding
the use in doing it. We hope that this will make it easier to
work with Konduit, which will thus appeal to more users.
The Konduit tool1for generating simple RDF workflows
(see Fig. 1) on the semantic desktop  is aimed at users
who are capable of using complex software, yet don’t have
any programming experience, or are simply not willing to
invest the time to program solutions for relatively simple,
but repetitive and time-consuming knowledge tasks. How-
ever, writing SPARQL queries — an important component
of Konduit — is still a considerable burden for those users if
it has to be done manually and might deter them from us-
ing Konduit. For this reason, we have built a first version of
a prototype to build SPARQL CONSTRUCT queries visu-
ally, thus hopefully improving the usability of Konduit sig-
nificantly. If the research into visual interfaces of SPARQL
queries turns out to be successful, we predict that those in-
terfaces will also be useful in contexts outside of Konduit.
In previous works  we have presented Konduit, a desktop-
based tool to visually build simple workflows with RDF data.
Konduit is integrated with the NEPOMUK KDE semantic
desktop environment . In NEPOMUK data from vari-
ous desktop applications — e.g., contact details, document
metadata or appointments and other events — as well as
arbitrary other metadata is interlinked and represented in
an integrated graph, and thus accessible through a uniform,
system-wide interface. With Konduit, a knowledge worker
can then use their desktop data, combine it freely with RDF
data available on the Web, filter, merge, trigger function-
ality of other desktop applications and integrate arbitrary
scripting code snippets into simple workflows which auto-
mate common, repetitive tasks. Examples for such tasks
are generating reports from combined online and desktop
Figure 1: Inferface of the Konduit tool
resources, sending emails with pictures of the last institute
party to all members of the institute, etc.
2.1SPARQL in Konduit
An important feature of Konduit is the possibility to de-
fine filter and transformer elements which modify and pro-
cess incoming RDF into the desired output RDF. Tech-
nically, these elements are implemented as CONSTRUCT
queries in the SPARQL2query language for RDF. In the
initial version of Konduit, those queries had to be entered
manually into a plain text box by the user (see Fig. 2).
The target user of Konduit is a knowledge worker with
no or only basic programming skills, but good overall com-
puter skills and a good understanding of the data they are
dealing with. In other words, we are not (yet) looking at
the proverbial “my mother”, but someone in the grey area
between expert and na¨ ıve user. For such users, we consider
defining a SPARQL query as feasible, but writing it manu-
ally is still tedious and error-prone. What is needed is a way
to aid and guide users in defining those queries. In a first
step, a syntactic support will already help semi-expert users
in building queries faster and with fewer errors. A second
step, which would abstract away from the actual syntax and
model the query on a higher level, would make the system
Figure 2: SPARQL CONSTRUCT element in Kon-
accessible to na¨ ıve users as well.
In this demo, we will present our current work in pro-
viding for basic syntactic support for query building. The
first iteration of the interface is very simple, yet already very
helpful in preventing errors and lowering the cognitive work-
load of the users. While the SPARQL text boxes from the
previous version of Konduit still exist, the user is now pre-
sented with an additional helper window, in which they can
assemble the query visually. From this interface, the query
in the text box is generated.
Both the CONSTRUCT and the WHERE block of the
query are built as series of triple patterns3, for which each
component is entered into a text field. When each text field
is filled, the pattern can be added to the list. This very
basic functionality is further extended by the application of
drop-down lists for predicates, as well as auto-completion.
The drop-down lists are assembled by previously inspecting
the current data in the NEPOMUK RDF store, while the
auto-completion takes its completion suggestions from both
the RDF store and the terms which have so far been entered
as part of the query. Additionally, the user can also specify
data types for literals from a selection box.
3.CONCLUSIONS AND OUTLOOK
The interface presented in this short paper is only a first
step to provide non-expert users with a way to define SPARQL
queries, and thus filter and manipulate data on the Seman-
tic Web (or Desktop). However, because it prevents the user
from errors, relieves them off typing effort and, according to
our intuition, lowers the learning curve required for building
queries, we think that an interface like this already provides
significant advantages over manual query writing.
For further research, we will be looking into abstract-
ing from the SPARQL syntax completely, and instead use
an interface that is based on the visual representation of
nodes and arcs instead, such as the“InteractiveSparqlQuery-
Builder”4by OpenLink Software. However, an interesting
research question would be to find how, for queries of differ-
ing complexity, a graph-based visualisation is actually more
or less intuitive. An alternative idea would be an interface
that applies the idea of Query by Example to SPARQL. Fi-
nally, a combination of approaches is worth investigating,
3The syntactic power of SPARQL is obviously restricted in
Figure 3: Building a SPARQL CONSTRUCT query
where parts of the query (a simple graph pattern) would be
represented graphically, whereas other aspects of SPARQL
would still be edited in the syntax, but with UI support.
The work presented in this paper was supported (in part) by
the L´ ıon project supported by Science Foundation Ireland un-
der Grant No. SFI/02/CE1/I131 and (in part) by the European
project NEPOMUK No FP6-027705.
 S. Decker and M. R. Frank. The networked semantic
desktop. In WWW Workshop on Application Design,
Development and Implementation Issues in the
Semantic Web, 2004.
 T. Groza, S. Handschuh, K. M¨ oller, G. Grimnes,
L. Sauermann, E. Minack, C. Mesnage, M. Jazayeri,
G. Reif, and R. Gudjonsdottir. The NEPOMUK
project — on the way to the social semantic desktop.
In T. Pellegrini and S. Schaffert, editors, Proceedings of
I-Semantics’ 07, pages pp. 201–211. JUCS, 2007.
 K. M¨ oller, S. Handschuh, S. Tr¨ ug, L. Josan, and
S. Decker. Demo: Visual programming for the semantic
desktop with Konduit. In S. Bechhofer, editor,
Proceedings of the 5th European Semantic Web
Conference (ESWC2008), Tenerife, Spain, volume 5021
of LNCS, pages 849–553. Springer, June 2008.