=Paper=
{{Paper
|id=Vol-1259/method2014_overview
|storemode=property
|title=Overview of METHOD 2014: the 3rd International Workshop on Methods for Establishing Trust of (Open) Data
|pdfUrl=https://ceur-ws.org/Vol-1259/method2014_overview.pdf
|volume=Vol-1259
|dblpUrl=https://dblp.org/rec/conf/semweb/NiesCGHM14
}}
==Overview of METHOD 2014: the 3rd International Workshop on Methods for Establishing Trust of (Open) Data==
https://ceur-ws.org/Vol-1259/method2014_overview.pdf
Overview of METHOD 2014: The 3rd
International Workshop on Methods for
Establishing Trust of (Open) Data
Tom De Nies1 , Davide Ceolin2 , Paul Groth2 , Olaf Hartig3 , and Stephen Marsh4
1
tom.denies@ugent.be
Ghent University – iMinds – Multimedia Lab, Belgium
2
d.ceolin, p.t.groth@vu.nl
VU University Amsterdam, The Netherlands
3
ohartig@uwaterloo.ca
University of Waterloo, Canada
4
stephen.marsh@uoit.ca
University of Ontario Institute of Technology, Canada
1 Introduction
The METHOD workshop aims to bring together researchers working on the
problem of trust and quality assessment of (open) data, and all components
that contribute to this goal. It provides a forum for researchers from both the
Semantic Web and the Trust Management community, with the goal of gaining
new insights towards solutions for this complex problem. Due to the relatively
low number of submissions, and to maximize the impact of the accepted papers,
METHOD 2014 merged with the 10th International Workshop on Uncertainty
Reasoning for the Semantic Web (URSW), as a special session. In this short
editorial paper, we provide an overview of the topics discussed during this session.
Trust assessment of content on the Web is a highly complex concept that
depends on objective as well as subjective criteria, including the content’s prove-
nance, data quality estimation, and also the consumer’s background, personality,
and context. However, the exact criteria and tolerances differ for each context
and for each assessor, requiring detailed knowledge about the data and its uses.
This also makes it very challenging to find generic solutions and assessments that
are applicable everywhere or transferrable from one context to another. There-
fore, stakeholders in this field are continuously investigating new techniques to
handle and prepare data in such a way that it becomes easier for machines to
process it with the goal of trust and/or quality assessment. The METHOD work-
shop is a venue for presenting and discussing novel research ideas in this field,
as well as technical applications.
2014 is the third year for METHOD. The two previous editions were held in
conjunction with the IEEE Annual International Computer Software & Appli-
cations Conference (COMPSAC). This year, the workshop is held for the first
time at ISWC, since the topics of provenance, data quality, and trust are highly
relevant to the Semantic Web community. The diversity in the ways that these
topics may be approached is also visible judging from the subject of the submis-
sions we received. Although only three papers were accepted, each submission
covers a distinctly different aspect of the general theme of the workshop.
2 Papers and Discussion Topics
This year’s submissions cover the following aspects of trust of (open) data: con-
tent rating, data quality rewarding, data attribution, and trust representation.
The first two aspects are covered by Couto, who proposes a number of guide-
lines for a system that rates as well as rewards good practices in data sharing,
by means of a virtual currency [1]. A large number of unsolved technical chal-
lenges are identified by this position paper, and some issues are left open, such
as how to uniquely identify and attribute data to its creators. This is exactly
what Höfig and Schieferdecker investigate, proposing a new hash function for
RDF graphs [3]. The solution proposed in this research paper may contribute
to a tamper-resistant way of attributing RDF data to its authors, allowing one
to make claims about the data’s trustworthiness. Of course, a representation
for these trustworthiness assessments is needed. In the final position paper of
our workshop, Ceolin et. al. propose an ontology to represent trust of web data,
extending existing solutions [2].
Despite the diverse aspects discussed in the submissions, a number of common
themes and open questions can be identified, listed below.
– What incentive is there for data creators to make their data trustworthy?
– Which mechanisms are in place to attribute data to its creators and editors,
and do they suffice for our needs?
– How can we estimate ratings or assessments of data quality and trustwor-
thiness?
– How would it be possible to allow the reuse of such trust and quality esti-
mations?
– How do we represent the aforementioned aspects in an interoperable way?
These are the questions we hope to see addressed during the discussions at
METHOD 2014, and in future editions of the workshop.
References
[1] Couto, F.: Rating, recognizing and rewarding metadata integration and sharing on
the semantic web. In: Proceedings of the 10th International Workshop on Uncer-
tainty Reasoning for the Semantic Web (URSW), Riva del Garda, Italy. CEUR-
WS.org (2014)
[2] Davide Ceolin, Archana Nottamkandath, W.F., Maccatrozzo, V.: Towards the def-
inition of an ontology for trust in (web) data. In: Proceedings of the 10th Interna-
tional Workshop on Uncertainty Reasoning for the Semantic Web (URSW), Riva
del Garda, Italy. CEUR-WS.org (2014)
[3] Hoefig, E., Schieferdecker, I.: Hashing of RDF graphs and a solution to the blank
node problem. In: Proceedings of the 10th International Workshop on Uncertainty
Reasoning for the Semantic Web (URSW), Riva del Garda, Italy. CEUR-WS.org
(2014)