The DBpedia dataset has multiple URIs within the dataset and from other datasets connected with (transitive) owl :sameAs relations and thus referring to the same concepts. With this heterogeneity of identi ers it is complicated for users and agents to nd the unique identi er which should be preferably used. We are introducing the concept of DBpedia Unique Identi er (DUI) and a dataset of linksets relating URIs to DUIs. In order to improve the quality of our dataset we developed a mechanism that allows the user to rate and suggest links. As proof of concept an implementation with a graphical web user interface is provided for accessing the linkset and rating the links. The DBpedia sameAs service is available at http://dbpsa.aksw.org/SameAsService.
Further DBpedia has more than one URI representing the same resource within the dataset, e.g. the dbpedia:Brassil4 has at least the following equivalents within the DBpedia dbpedia:Republica_Federativa_do_Brasil, dbpedia:ISO_ 3166-1:BR and dbpedia:Brazil which are all redirecting to dbpedia:Brazil. Thus a problem to consider is to directly resolve any of the equivalents directly to the nal URI e.g. http://dbpedia.org/resource/Brazil without any redundancies.
Also, according to Halpin et. al. [1] and Wood et. al. [4], sameas.org has collected millions of triples with owl:same As relations. It would be important to promote reciprocal owl:sameAs con rmation mechanisms and develop e ective trust mechanisms to assure the quality of owl:sameAs relations.
To tackle the identi er heterogeneity problem we are making the following contributions:
We describe an approach for the mitigation of the identi er heterogeneity problem and implement a prototype where the user is able to evaluate existing links, as well as suggest new links to be rated.
The ability to generate statistics about good and bad links which, brings the possibility to have a quality control for the links to DBpedia.
We de ne the DBpedia Unique Identi er (DUI), which instead of several transient owl:sameAs DBpedia URIs for the same nal address, now is possible to have a unique URI from DBpedia. A DUI goes directly to the nal address instead of having to process several possible intermediate results. For example, with a URI from Freebase, 17 redundant URIs from DBpedia where avoided or if one used a service such as sameAs.org, 1141 URIs would be avoided.
The rest of the paper is organized as follows: section 2 represents a proposed approach for tackling the identi er heterogeneity problem, we evaluate our work in section 3, in section 4 we focus on related work, and nally section 5 concludes the paper and outline future work. 4Throughout the paper we are using the following namespace de nitions: owl: http://www.w3.org/2002/07/owl#, dbpedia: http://dbpedia.org/resource/.
This section provides an explanation about our main idea, such as implementation and descriptions.
Before continuing the work, there are some de nitions that were adopted.
Normalization of the URI: Is understood by normalizing URIs, the fact of eliminating redundancies. DBpedia unique identi er: The DBpedia Unique Identi er (DUI) is an unique URI that identi es a resource in the DBpedia repository and also is the result of our normalization.
The idea started with a stand alone service on the web that solves the problem where the user provides a URI as parameter and instead of several transient URIs with owl:sameAs property, the user receives a single DUI from our service.
The work- ow for requesting the DUI of a given resource is represented in g. 1. Firstly, the user will provide a URI from some address, i.e. FreeBase. Then, instead of possible several results of URIs with the property owl:sameAs, our system will return a DUI. Consequently, the user has a possibility to rate, verify, validate, and suggest a di erent link. Then the rate can give us a chance to have statistics about the quality of the links.
A service, also was implemented, where the user can provide a URI and the API will return the DBpedia identi er like a URI that represents the owl:sameAs about the URI provided.
This section describes in four steps the technique and how
the idea was developed, from phase of importing links to a
relational database until the development of the service on
the web and a GUI.
(
Where the DBpedia Link Repository uses the DBpediaSameAs service in order to tackle the heterogeneity and giving the appropriate DUI, that redirects the user to the DBpedia Link Rate interface, thus, providing a feedback to the DBpedia Link Repository, therefore, improving the quality of the DBpedia endpoint.
The aim of this qualitative6 evaluation was centered in verifying the behavior of the service DBpediaSameAs, the Graphical User Interface (GUI) that gives the possibility to verify and rate the links.
There are chosen 3 evaluation criteria:
(
The criteria used in this evaluation are uniquely to tackle heterogeneity, that was observed during the search of coreferences between di erent data sets with a problem about redundancies.
When was used a URI from freebase in order to obtain a DBpedia URI was observed that at least 3 URIs were returned, that drives to the same nal address.
As an example of a real case, executed in our public server, with a URI from Freebase: $ curl http :// dbpsa . aksw . org / :
S a m e A s S e r v i c e / S a m e A s S e r v l e t ? uris = http %3 : A %2 F %2 Frdf . freebase . com %2 Fns %2 Fm .015 fr returns : http :// dbpedia . org / resource / :
Brazil Where, in this case, instead of 17 URIs from DBpedia, that goes to the same nal address, our approach drives the user directly to the nal address.
As can be observed on the gure 4 that approach the transitive and redirect URIs, where show that with this approach instead of have several URIs the user can have only one from the DBpediaSameAs. Thus, in this way, providing a normalization on DBpedia URIs.
In order to have a link rating, were implemented a GUI that allows the users to give some feedback, suggestions, in this way, improving the quality of the links. The rate is a quite simple process, the GUI just ask the user to rate the link with +1 if the link attends your expectations or -1 if the link is wrong or some type of spam. The GUI was developed using concepts from pre x.cc 7 and work from Zaveri[5] such our system of rate (+1 and -1) and the standard of the web documents. Some improvements and personalization, also was provided, such as the suggestions and the possibility to check the link. The gure 3 shows the moment when the user clicked on the -1 and indicated that the user didn't like the link and was asked to make a suggestion of a new URI. The eld about a suggestion for a new link will only appear when the user are not satis ed with the current link, then, when clicking on the -1, then the system will ask for an optional suggestion.
The results of this work could also be expressed in numbers that was obtained during importing triples to the relational database and with some results from the sameAs.org web site. A total of 62,531,487 triples imported into our database, the time was 2,220 seconds for the whole operation, thus, was noticed that 28,167 triples were imported per second. The source code used to obtain the results is available in our github repository8. 7http://prefix.cc 8https://github.com/firmao/dbpedia-links/blob /master/CreateDB.sh 3.3.1 Transitive and Redirect Links Transitive and Redirect Links are redundancies at DBpedia that supposed has a link to the same place, in other words, they use owl:sameAs property, this links will redirect another links, will provide a transition between the links, that's why the name transitive. In this case, instead of using this transitive links that points to the same nal destination URI, this nal destination URI will be used directly. The gure 4 try to make more clear this explanation.
Was discovered and treated 6,473,988 triples with transitive and redirect links from 62,531,487 imported links among 142 domains inside DBpedia. Then, 10.35% of the links can be avoided in some cases.
The DBpediaSameAs was evaluated with its normalization of URIs, link rate, and DBpediaSameAs as a stand alone service on the web. As results of the normalization a DUI was obtained in order to tackle the heterogeneity. In other words, instead of several URIs e.g. from sameAs.org one DUI was obtained. The link rate functionality further allows to improve the quality of the dataset.
Despite, the GUI of DBpediaSameAs, also a stand alone service on the web was developed that brings the functionality to get a DUI without a GUI for agents or people which don't need to use the DBpediaSameAs in a Graphical mode, allowing use as an o -the-shelf component.
The work [5] elaborates a data quality assessment methodology in DBpedia, which comprises of a manual and semiautomatic process. This work drive us to a reinforcement about the concept of data quality used in our work, when in our case will be more a manual process and also we are able to improve the DBpedia data quality.
The work [2], presents a two staged experiment for the creation of gold standards that act as benchmarks for several interlinking algorithms. The similar aspects of this works are: The validation of links and a dubbed manual validation, where the user i.e. validator or evaluator speci es whether a link generated by an interlinked tool is correct or incorrect. The results of the link validation process are used to learn presumably better link speci cations and thus achieving high-quality. Also, this work proposes an experiment to investigate the e ect of user intervention in dataset interlinking on small knowledge bases.
The sameAs.org is a service that leading source of co-reference data on the Semantic Web. For example, when the web site sameAs.org is accessed with a URI from Freebase that should bring information about a country called Brazil. Was used the URI (http://rdf.freebase.com/ns/m.015fr) as parameter to the service, and is received as return more than 1140 URIs as shown in g. 5, but the user can have a doubt about which one is the correct.
Our work is not an alternative to the sameAs web site, but brings possibilities, like, was noticed that the sameAs.org does not provide a way to rate the link, but with this rating, is possible to improve the quality of the data, and bring some facility to the user.
An approach was provided to tackle the heterogeneity working with owl:sameAs redundancies that were observed during researching co-references between di erent data sets and providing a unique DBpedia identi er and give the chance to rate the resulting links and make suggestions. A proof of concept was implemented as a computer web system in order to present and validate our idea and every concept of this work. The source code is available 9. Was noticed in our results that there are bene ts when a considerable number of owl:sameAs redundancies can be avoided. Rating the links allow users to make link suggestions brings more quality to the repository, and the stand alone service on the web allow you to use the DBpediaSameAs also in a command line textual environment and can be used as an o -the-shelf component.
For the future we plan to: (
We would like to acknowledge, National Council for Scienti c and Technological Development (CNPq) 10 and Universitat Leipzig for their support. Special thanks to Markus
Ackermann for a essential help with the deployment of application and very good suggestions. Additionally,research activities of this paper were funded by grants from the EU's 7th & H2020 Programmes for projects ALIGNED (GA 644055), GeoKnow (GA 318159) and LIDER (GA 610782).