Monthly Archives: April 2011

Pascal Hitzler

Semantic Web and Emerging Trends in Scholarly Publishing

by , Category: Companies & Institutions, Conferences & Events, Semantic Web Applications, Uncategorized

In my capacity as one of the Editors-in-chief of the Semantic Web journal (the other one is Krzysztof Janowicz; the journal is published by IOS Press), I was recently invited to talk about the journal at Allen Press’ Seminar Emerging Trends in Scholarly Publishing.  This seminar is an annual event which draws decision makers from the scholarly publishing industry to hear about and discuss recent developments and hot topics related to their profession. This year’s event had a session on “Semantic Enrichment”, and one on “Rethinking the Structure of Peer Review.” All presentations, including videos, are available from the Allen Press website.

The invited speaker of the “Semantic Enrichment” session was Pam Harley, Vice President, Product & Market Development of Semedica, a division of Silverchair.  Pam gave a high-level account of the possibilities and added value which comes with Semantic Enrichment, in a way suitable for the non-technical audience. I personally benefited particularly from the large variety of reasons for adopting Semantic Technologies in publishing which she presented and discussed in her talk (see also her slides).

My presentation (see also the slides) about the Semantic Web journal was part of the “Rethinking the Structure of Peer Review” session, and was focused on the open and transparent review process which we have adopted for the journal. After the presentation, throughout the event, I received ample feedback and remarks which in particular commended us for setting up a realistic improvement of the review process while avoiding radical changes which are likely to meet too much resistance from researchers. I certainly agree with this assessment. The presentation also contains a bit of information on how the journal is doing (in short: it’s doing great).

The seminar was a very enjoyable experience. In particular, it was enlightening to learn about publisher’s perspectives on scientific publishing, reviewing processes, and emerging revenue models. It was also nice to see that Semantic Web as a technology has a natural place in these discussions and is seeing more and more adoption in practice.

If you’re curious to learn more, have a look at the videos of the presentations.

[Author: Pascal Hitzler]

Thomas Thurner

The hype, the hope and the LOD2: Sören Auer engaged in the next generation LOD

by , Category: Linked Data & Open Data, Tools & Software

The paneuropean Project LOD2 is one of the biggest projects dealing with linked data. Scientists, programmers and software architects in various european countries are working on the next generation of linked open data. In a series of interviews i’m presenting people working on and with LOD2. As a start, i had the change to talk to Sören Auer, head of the LOD2 project.

Thomas Thurner: Over the recent years the LOD movement gained tremendous momentum. As one of the key players in this area how do you perceive this development? Hype or hope?

Sören Auer: From my point of view the momentum LOD gained is deserved. We should strive for a Web, which is more decentralized, democratic, participatory, transparent and inclusive. Linked Open Data is from my point a key technological building block on this road. However, a lot of work is ahead of us. LOD has to find its way directly into mainstream technology such as CMSes, Search Engines, Web Applications, Mash-Ups and we have to show users and stakeholders the direct added-value of this technology.

Thomas Thurner: What is the current state of the LOD cloud from a technological point of view? Where do you see room for improvement?

Sören Auer: Currently, the technological state of LOD seems to be comparable to the early days of the Web. We are still able to draw maps/clouds of the LOD datasets and data links are still sparse and difficult to maintain. This reminds me a lot of the early days of the Web, where we also had problems with broken links (the infamous 404). Later, after content management systems and Web applications automatized the link generation and maintenance this improved a lot and I hope we are on the same road with LOD technologies finding its way into more and more Web systems.

Thomas Thurner: How is the LOD2 project addressing theses issues? What are the project’s key objectives?

Sören Auer: LOD2 is addressing in three ways: First, we develop new research approaches highly relevant for LOD, for example, for Linked Data management, automatic data linking as well as Linked Data enrichment andquality improvement. Second, we implement and integrate these approaches into specialized tools (e.g. SILK, OntoWiki, Virtuoso and DL-Learner) forming together the integrated LOD2 stack. The LOD2 stack can be used by data publishers for the whole life-cycle of Linked Data management ranging from extraction over linking, authoring, enrichment to exploration & search.

Thomas Thurner: What do you think are the most important factors to bring LOD to the masses?

Sören Auer: From my point of view the key factor here is that we manage to integrate the large number of tools and approaches for supporting the Linked Datalife-cycle stages in a synergistic way, where each aspect adds value and triggers a number of other improvements. For example, the establishing of a new data link has a direct effect on search & exploration of Linked Data. We have to directly show these kind of benefits to users so they receive and instant gratification for contributions to the Web of Data. Semantic Wikis, such as Semantic MediaWiki and OntoWiki, are already nicely working in this direction. An application with an enormous potential to bring LOD to the masses would be the creation of a distributed, social semantic network. With OpenId, WebId, FOAF, Semantic Pingback most of the building blocks are available, but the final step integrating these into an easy-to-use social networking application still has to be done.

Thomas Thurner: Compared to other semantic web approaches linked data principles seem to be rather easy to understand. On the other hand some argue that the “linked data cloud” is a big heap of data which cannot be used for professional purposes. What is your point of view?

Sören Auer: Of course the currently available data is not useful for all potential usage scenarios. However, already now Linked Data can be used for many interesting applications: For example, we just completed the development of a prototype for a large search engine, where users searching are assisted with comprehensive background information obtained from the Linked Data Web. For this use case, information available as Linked Data is already very valuable and useful. The criticism of LOD being a “heap of data” also reminds me a lot of the early days of the Web, where people raised similar criticisms for the Web being a medium of un-professionalism. Later it turned out that, of course there is a lot of amateurism, but as Wikipedia impressively demonstrates the working together of many amateurs with the right tools can in the end outperform few professionals.

Thomas Thurner: Linked Data could also become a new paradigm for light-weight enterprise data integration. What are the biggest obstacles today for linked data to being accepted by the business community?

Sören Auer: Using Linked Data for data integration in large enterprises has an enormous potential. Just last week I was invited for a workshop with the IT department of one of the top car makers and the people responsible there for data integration were extremely excited about the opportunities of Linked Data in the large heterogeneous enterprise with more than 3000 different backend systems. Linked Data technologies can easily fill the gap between unstructured Intranet search and expensive & complicated Service-oriented Architectures. Compared to SOA, Linked Data is a pay-as-you-go strategy, where data integration can be performed incementally and in sync with the requirements and evolution of the data structures in the enterprise. In order to realize this vision, we need to continue the maturation of enterprise Linked Data tools – the availability of PoolParty, Sindice Enterprise Edition, Virtuoso, TopBraid are already important steps in that direction.

Thomas Thurner: Automatic mechanisms to curate linked data and to make alignments between datasets possible play a crucial role for the next phase of linked data economics. Which technologies will play a central role? What will be the most critical point – do you see a “wisdom of the crowd” playing a role in this game?

Sören Auer: Definitely! Tapping the wisdom of the crowd for mapping & linking has a huge potential, which is currently unused. We started working in that direction with DBpedia Live and the DBpedia mapping Wiki. In order, to make it really easy for people to contribute we have to dramatically lower the barrier to contributing to the alignment process. In LOD2 we also plan to enable users to create mapping and links between dataset by simply giving examples of correct links and evaluating some automatically generated ones.

Thomas Thurner: At the moment governments all around the world start to publish open data, more and more stakeholders start to understand the benefit of open linked data. On the other hand enterprises haven´t even started with this topic. What could be the dynamics which will trigger projects in industry sectors like financial industries which will make use of open data principles?

Sören Auer: Making statistical and financial information available in structured form and as Linked Data could have a enormous impact in this regard. With the DataCube vocabulary effort a first step in this direction was made, but it would be nice if this vocabulary would get an official stamp of a standardization organization such as W3C. Since the benefit of publishing statistical and financial data in structured form, e.g. as Linked Data, is visible most when done by many, this could be also facilitated by government regulations and industry best-practices.

About INFAI

The Institute for Applied Computer Science (InfAI) at Universität Leipzig hosts research groups in service sciences, knowledge engineering and management as well as natural language processing. The approximately 20 researchers of the Agile Knowledge Engineering and Semantic Web (AKSW) research group at InfAI headed by Dr. Sören Auer are establishing theoretical results and scalable implementations for the field. Particular emphasis is given to areas such as ontology creation and
manipulation, knowledge extraction, ontology learning and information & data integration on the Semantic Data Web. The implemented tools and services (such as DBpedia, OntoWiki, DL-Learner and LinkedGeoData) developed by the group enjoy considerable popularity.

About Sören Auer

Dr. Sören Auer leads the research group Agile Knowledge Engineering and Semantic Web (AKSW) at Universität Leipzig. His research interests include semantic data web technologies, knowledge representation, engineering & management, usability, agile methodologies as well as databases and information systems. He aims to combine strong theoretical results with high-impact practical applications. Sören is author of over 50 peer-reviewed scientific publications resulting in a Hirsch index of 15. Sören is leading the large-scale integrated EU-FP7-ICT research project “LOD2 – Creating Knowledge out of Interlinked Data”. Sören is founder (respectively co-founder) of several high-impact research and community projects such as the Wikipedia semantification project DBpedia or the social Semantic Web toolkit OntoWiki. He is co-organiser of several workshops, programme chair of I-Semantics 2008, OKCON 2010, ESWC 2010 and ICWE 2011, area editor of the Semantic Web Journal, serves as an expert for industry, the European Commission, the W3C and is member of the advisory board of the Open Knowledge Foundation.

Andreas Blumauer

Controlled vocabularies: “Data integration is king”

by , Category: Calls & Competitions, Vocabularies & Languages

Just recently a survey about “Controlled vocabularies” and their significance for enterprise information management has started. Until today 143 participants have responded and completed the survey at least partially. To give a first example what was found out, I would like to take a closer at the question: What are the main application areas of controlled vocabularies from your perspective?

A bit surprising is the intermediate result, that it´s not “Semantic Search” or “Support of multilingual applications” which was considered to be the most important application. Instead of this it turned out that “Data Integration” is king:



The bar graph shows the weighed value of each application candidate (1.0 would be a 100% acceptance that this is an important application area of controlled vocabularies). Regarding the top candidate “data integration”

  • 57,4% said “very important”
  • 29,8% “relevant”
  • 7,4% “somewhat relevant”
  • 2,1% “not relevant”
  • 3,2% “Don´t know”

If you don´t think this should be the final result, please help to get a better overview of what´s going on in the controlled vocabulary community. The survey is open until May 18th, 2011 – all participants will gain access to a report with the results within the following month. Most interesting results will be made public on this blog.

Thomas Schandl

Linked data based thesaurus management in collaborative settings

by , Category: Corporate Semantic Web, Enterprise 2.0, Knowledge Management, Linked Data & Open Data, Semantic Web Applications, Tools & Software

The creation and management of controlled vocabularies in companies often takes place in a distributed manner. Different departments in different branch offices often rather create their own vocabularies, than have one large central knowledge model, where everyone contributes.

How to model divergent views on one concept?

Such a central model is not only much harder to manage, but there is also the general problem that differerent departments like marketing, quality assurance, R&D, etc. will have divergent views on the model and its concepts. These different perspectives on one and the same concept are hard to unify in a single model.

Think of a company that sells mobile phones and wants to create a model of its line of products. It wants to utilize this model in the context of its online shop as well as in the context of its user support forum. While the structure of the model (i.e. the relationships between the products) might be very similar or the same in both contexts, there will be differences in which properties of the products are actually relevant in the respective contexts.

In the model of the marketing department there might be a concept for a “Phantastax StamiMaxx” cell phone with a definiton “The StamiMaxx has a powerful battery and is great for professionals who travel a lot”. They might relate it to manufacturer “ACME Corporation” and to several concepts representing different features like “Android OS”, “Multi-touch touchscreen”, etc.
The very same phone has different properties that are interesting from the Quality Assurance departement’s perspective. They might call it by a more specific name like “Phantastax i3000 StamiMaxx S”, have a different definition for it like “3G cell phone implementing the new WTF3000 protocol, …” and relate it to concepts representing known problems and their solutions.

Now they face the task to integrate these different models, as it is not desirable to use a bunch of isolated models within one company.

Support of collaborative work on distributed models

To support this kind of collaborative work on distributed knowledge models, we would like to link the concepts of the models, just as is we link documents in the World Wide Web. Fortunately the Simple Knowledge Organisation System (SKOS) offers mapping properties that can be used to define relationships between concepts from different knowledge models.

E.g. when we want to say that concept “Phantastax StamiMaxx” in the product line thesaurus refers to the same real world entity as concept “Phantastax i3000 StamiMaxx S” in the Quality Assurance thesaurus, then we can use skos:exactMatch to express that. If we want to express that the concepts are merly similar, skos:closeMatch could be used.

The other SKOS mapping properties express a hierarchical (narrowMatch, broadMatch) or an associative (relatedMatch) mapping relation between concepts from different concept schemes. With those we can say that my Samsung Galaxy concept has a skos:broadMatch “Smartphone” in the product line vocabulary and a skos:relatedMatch “ACME Corporation” in a controlled vocabulary about Tech companies.

Modularisation of knowledge models

In this way SKOS thesaurus management systems like PoolParty make it possible to modularise knowledge models, represent concepts in their different contexts and consequently enable collaborative work on those models: The marketing guy can work on his model with the concept properties focused on sales without disrupting the work of the quality assurance expert on her own thesaurus. Later one or both of them can create the skos:exactMatch link between the concepts that are the same, like seen in the “Exact Matching Concepts” box in screenshot of PoolParty below.

Enrich your knowledge: Get connected with the LOD Cloud

Going a step further the models could be connected to external knowledge, e.g. a source from the Linked Open Data (LOD) Cloud. Once we establish links to LOD hubs like DBpedia, we can import additional information for their concepts or use it to establish whether similar concepts from different models really refer to the same real world resource.