Preface



Linked Data have attracted a lot of attention from both developers and researchers in re-
cent years, as the underlying technologies and principles provide new ways, following the
Semantic Web standards, to overcome typical data management and consumption issues
such as reliability, heterogeneity, provenance or completeness. Many areas of research
have adopted these principles both for the management and dissemination of their own
data and for the combined reuse of external data sources. However, the way in which
Linked Data can be applicable and beneficial to the Knowledge Discovery in Databases
(KDD) process is still not completely understood.
The Linked Data 4 Knowledge Discovery workshop (LD4KD), co-located within the
ECML/PKDD2014 conference in Nancy (France), explores the benefits of Linked Data
for the very well established KDD field. Beyond addressing the traditional data manage-
ment and consumption KDD issues from a Semantic Web perspective, the workshop aims
at revealing new challenges that can emerge from joining the two fields.
In order to create opportunities for communication as well as collaboration channels, the
workshop accepted 8 research papers from practitioners of both fields. The first observa-
tion one can make from those contributions is that the most obvious scenario for using
Linked Data in a Knowledge Discovery process is the representation of the underlying
data following Semantic Web standards, as shown in [De Clercq et al.], [Bloem et al.] and
[Krompass et al.], with the aim of simplifying the knowledge extraction process. With that
said, other contributions targeted other aspects of KDD, such as data pre-processing or
pattern interpretation, with the purpose of showing that KDD processes can benefit from
including elements of Linked Data. For the purpose of data preparation, [Rabatel et al.]
focuses on mining Linked Data sources, while [Zogała-Siudem et al., Ristoski et al.] use
Linked Data to enrich and integrate local data. The interpretation step of the KDD process
is also addressed, in the work of [Alam et al.] on results interpretation and the one of
[Peña et al.] on visualisation. We thank the authors for their submissions and the program
committee for their hard work. We sincerely hope that this joint work will provide new
ideas for interactions between those two, mostly isolated communities.


                                                                             Ilaria Tiddi
                                                                         Mathieu d’Aquin
September 2014                                                               Nicolas Jay