=Paper=
{{Paper
|id=None
|storemode=property
|title=Linked Soccer Data
|pdfUrl=https://ceur-ws.org/Vol-1026/paper6.pdf
|volume=Vol-1026
|dblpUrl=https://dblp.org/rec/conf/i-semantics/BergmannBEHKSS13
}}
==Linked Soccer Data==
https://ceur-ws.org/Vol-1026/paper6.pdf
Linked Soccer Data
Tanja Bergmann1 , Stefan Bunk1 , Johannes Eschrig1 , Christian Hentschel2 ,
Magnus Knuth2 , Harald Sack2 , and Ricarda Schüler1
1
firstname.lastname@student.hpi.uni-potsdam.de
2
firstname.lastname@hpi.uni-potsdam.de
Hasso Plattner Institute for Software Systems Engineering, Potsdam, Germany
Abstract. The sport domain is strongly under-represented in the Linked
Open Data Cloud, whereas sport competition results can be linked to al-
ready existing entities, such as events, teams, players, and more. The
provision of Linked Data about sporting results enables extensive statis-
tics, while connections to further datasets allow enhanced and sophisti-
cated analyses. Moreover, providing sports data as Linked Open Data
may promote new applications, which are currently impossible due to
the locked nature of today’s proprietary sports databases. We present a
dataset containing information about soccer matches, teams, players and
so forth crawled from from heterogeneous sources and linked to related
entities from the LOD cloud. To enable exploration and to illustrate
the capabilities of the dataset a web interface is introduced providing a
structured overview and extensive statistics.
Keywords: Linked Data, Soccer, Information Extraction, Triplification
1 Introduction
The Linked Open Data (LOD) Cloud includes 870 datasets containing more
than 62 billion triples1 . The majority of triples describes governmental (42 %)
and geographic data (19 %), whereas Linked Data about sports is strongly under-
represented. Sport competition results are collected by various authorities and
other parties, they are connected to events, teams, players, etc. Providing also
Linked Data about sports and sporting results enables extensive statistics, while
connections to further datasets allow enhanced and sophisticated analyses. More-
over, providing sports data as Linked Open Data may promote new applications,
which are currently impossible due to the locked nature of today’s proprietary
sports databases. By enabling linkage to additional resources such as geographi-
cal, weather, or social network data, interesting statistics for the sport enthusiast
can be easily derived and provide further information that would be hidden oth-
erwise.
In this paper we describe an extensive RDF dataset of soccer data provid-
ing soccer matches, teams, and player information, collected from heterogeneous
1
http://stats.lod2.eu/
25
Linked Soccer Data
sources and linked to LOD datasets like the DBpedia. The raw data was collected
via APIs and crawling from authorities’ websites, like UEFA.com or Fussball-
daten.de, and is linked to further web resources for supportive information, such
as Twitter postings for most recent information, Youtube videos for multime-
dia support, and weather information. Based on this aggregated new dataset we
have implemented an interactive interface to explore this data.
2 Related Work
The BBC Future Media and Technology department applies semantic technolo-
gies according to their Dynamic Semantic Publishing (DSP) strategy [2] to au-
tomate the publication, aggregation, and re-purposing of inter-related content
objects. The first launch using DSP was the BBC Sport FIFA World Cup 2010
website2 featuring more than 700 team, group and player pages. But, the data
used by the system internally is not published as Linked Data.
An extensive dataset of soccer data is aggregated by footytube. According to
their website3 the data is crawled from various sources and connected by seman-
tic technologies, though the recipes are not described in detail. Footytube’s data
include soccer statistics about soccer matches and teams, as well as related me-
dia content, such as videos, news, podcasts, and blogs. The data is accessible via
the openfooty API but is subject to restrictions that interdict the re-publishing
as Linked Data.
Generally, it is hard to find open data about sport results, since exploita-
tion rights are possessed by responsible administrative body organizations. An
approach to liberate sport results are community-based efforts, such as Open-
LigaDB4 , which collect sport data for public use. Van Oorschot aims to extract
in-game events from Twitter [3]. As to the authors’ best knowledge, the pre-
sented dataset provides the first extensive soccer dataset published as Linked
Data, consisting of more than 9 million triples.
3 Linked Soccer Dataset
Our intention was to create a dataset including reliable information about soccer
events covering as many historical data as available including recent competition
results. For this purpose DBpedia as cross domain dataset is not sufficient, since
soccer data in DBpedia is incomplete and unreliable.
The dataset is aggregated from raw data originating from Fussballdaten.de5 ,
Uefa.com6 , DBpedia7 , the Twitter feed of the Kicker magazine8 , the Sky Sport
2
http://news.bbc.co.uk/sport2/hi/football/world_cup_2010/default.stm
3
http://www.footytube.com/aboutus/search-technology.php
4
http://www.openligadb.de/
5
http://www.fussballdaten.de/
6
http://www.uefa.com/
7
the original http://dbpedia.org/ and German DBpedia http://de.dbpedia.org/
have been applied for matching
8
http://twitter.com/kicker_bl_li
26
Linked Soccer Data
HD Youtube Channel9 , and weather information from Deutscher Wetterdienst10 .
Fussballdaten.de, Uefa.com, and Kicker.de offer match results and player in-
formation. The Twitter feed is used both for parsing live match data (Kicker
updates its feed with live results) and to analyse free text tweets for latest
news about players or teams. The time frame of our data collection ranges from
the 1960s until today and is updated constantly. Updates are scheduled every
matchday, while the Twitter feeds are refreshed every 30 seconds during run-
ning games. Additional leagues can be included by setting up new crawlers, or
by providing an interface for manual submission. Currently, the dataset contains
information about 1. and 2. Bundesliga, the Champions League, European and
World Championships.
The data from these sources is converted and persistently stored as RDF
triples describing resources such as soccer player, soccer teams, matches, associ-
ations, different types of in-game-events, and seasons. Each entity is referenced
by a unique URI, which unites all facts, from whatever source they originate,
about the entity.
For describing the information about soccer data we have created a vocabu-
lary Soccer Voc 11 , which extends the BBC Sport Ontology[1] with soccer specific
classes and properties.
The dataset comprises descriptions of about 57,000 soccer players, 1,500
teams, 1,400 clubs, 1,500 referees, 1,800 managers, 700 stadiums, 38,000 matches,
97,000 goals, and 207 seasons or competition series. In total 9 million triples have
been generated up to now. About 3.35 million triples originate from raw data
from Fussballdaten.de and 2.10 million triples from the UEFA.com website.
In order to evaluate the quality of the matching, a percentage of matched
entities has been reviewed. The correctness of these matches was confirmed by
manually comparing the results to a data sample. For Bundesliga, all teams (54)
and about 78 % of all players (6,790) have been matched successfully to DBpedia
entities. Missing matches were mostly due to missing player entities in DBpedia.
4 Application
The soccer dataset comprises a diverse amount of information, both historic
and present data. As the data set contains data about every match played, it is
possible to create queries for all types of entities in a soccer match, e. g. all games
of a particular referee, or all games played in a specific stadium. By querying the
data, the user can find interesting statistics about the world of soccer, or find
information about his or hers favorite club.
The dataset can be accessed via a demonstrator website12 , where each entity
is presented on its own page with relevant information, statistics, and links to
9
http://www.youtube.com/user/SkySportHD
10
http://www.dwd.de/
11
http://purl.org/hpi/soccer-voc/
12
http://mediaglobe.yovisto.com/SoccerLD/
27
Linked Soccer Data
related entities. Additionally, a variety of possible complex queries are demon-
strated, such as “Which player is most important for his team?”, “From which
foreign country do most players in the last Bundesliga season come from?”, or
“Which team performs best in rainy weather?”. In Figure 1, two different views
of the website are shown.
Fig. 1. Left: Information about a German soccer club, among other a graph showing
promotions and relegations (generated from match data) and free text tweets belong
to this club, Right: Map visualization about the distribution of international players
in the Bundesliga since 1963, generated from player data.
5 Conclusion and Outlook
We presented a rich soccer dataset, which is to our best knowledge the first
comprehensive linked soccer dataset. We published non-restricted parts of the
dataset, the publication of the dataset as a whole is prevented by legal rights
belonging to the respective authorities. Applications based on this data not only
allow for typical statistical information about players and matches but also ex-
ploit the advantages of Linked Data principles in order to provide additional
information currently not considered by available soccer datasets. We have de-
veloped and deployed a website in order to conveniently browse the dataset and
provide various statistics that exemplify the advantage of aggregating multiple
resources as Linked Data.
Possible additions could include advanced and more detailed data such as
the number of ball contacts, played passes, or the distance covered by a player
during a match. Integrating such data even more sophisticated queries could
be answered. Further extensions of the dataset include also articles from sport
magazines like interviews, team presentations, or background stories of players.
28
Linked Soccer Data
References
1. S. Oliver. Enhancing the BBC’s world cup coverage with an ontology driven infor-
mation architecture. In P. F. Patel-Schneider, Y. Pan, P. Hitzler, P. Mika, L. Zhang,
J. Z. Pan, I. Horrocks, and B. Glimm, editors, 9th International Semantic Web Con-
ference (ISWC2010), November 2010.
2. J. Rayfield. Dynamic semantic publishing. In W. Maass and T. Kowatsch, edi-
tors, Semantic Technologies in Content Management Systems, pages 49–64. Springer
Berlin Heidelberg, 2012.
3. G. van Oorschot, M. van Erp, and C. Dijkshoorn. Automatic extraction of soccer
game events from twitter. In M. van Erp, L. Hollink, W. R. van Hage, R. Troncy, and
D. A. Shamma, editors, Proceedings of the Workhop on Detection, Representation,
and Exploitation of Events in the Semantic Web (DeRiVE 2012), volume 902, pages
21–30, Boston, USA, 11 2012. CEUR.
29