University of Leipzig AKSW Homepage | Blog |

Archiv für die Kategorie 'LIMES'

Five AKSW Papers at ESWC 2014

March 14, 2014 - 1:37 am by AxelNgonga - No comments »
Hello World!
We are very pleased to announce that five of our papers were accepted for presentation at ESWC 2014. These papers range from natural-language processing to the acquisition of temporal data. In more detail, we will present the following papers:
  • NLP Data Cleansing Based on Linguistic Ontology Constraints (Dimitris Kontokostas, Martin Brummer, Sebastian Hellmann, Jens Lehmann and Lazaros Ioannidis)
  • Unsupervised Link Discovery Through Knowledge Base Repair (Axel-Cyrille Ngonga Ngomo, Mohamed Sherif and Klaus Lyko)
  • conTEXT – Lightweight Text Analytics using Linked Data (Ali Khalili, Sören Auer and Axel-Cyrille Ngonga Ngomo)
  • HiBISCuS: Hypergraph-Based Source Selection for SPARQL Endpoint Federation (Muhammad Saleem and Axel-Cyrille Ngonga Ngomo)
  • Hybrid Acquisition of Temporal Scopes for RDF Data (Anisa Rula, Matteo Palmonari, Axel-Cyrille Ngonga Ngomo, Daniel Gerber, Jens Lehmann and Lorenz Bühmann)
Come over to ESWC and enjoy the talks. More information on these publications at
Axel on behalf of AKSW

Two AKSW Papers at ESWC

February 25, 2013 - 5:28 pm by AxelNgonga - No comments »

Greetings World.

We are happy to announce that the AKSW papers “COALA – Correlation-Aware Active Learning of Link Specifications” and “When to Reach for the Cloud: Using Parallel Hardware for Link Discovery” were selected for presentation at ESWC 2013. Each of the papers deals with one of the two main hurdles to scalable Link Discovery. COALA presents approaches that improve the convergence of active learning algorithms for learning link specifications. For this aim, COALA uses graph clustering and weight decay to include information on the distribution of most informative positive and negative link candicates in the selection process. The second paper deals with the scalability of Link Discovery and presents practical suggestions on when to use which hardware for linking. As a highlight, it presents GPU and Cloud implementations of the HR3 algorithm that we presented at ISWC 2012. All algorithms are implemented within the LIMES framework.

Looking forward to seeing you at ESWC and link on!

Axel for AKSW

More than 20 European Union Datasets Converted to RDF by LATC Project

July 9, 2012 - 2:13 pm by Jens Lehmann - 2 comments »

Over the past two years, the LATC project (Linked Open Data Around-The-Clock) has worked on converting more than 20 EU datasets to RDF, make them available as Linked Data and SPARQL, and link them to other datasets. The datasets have gone through internal quality assurance against a publication checklist. LATC supports the EU Open Data Portal initiative and aims to bootstrap the creation of an EU dataset cloud by this effort. The datasets include several business, legal and statistical datasets such as CORDIS, the European Commission Financial Transparency System, EUR-Lex, EuroStat and many more. A full list of all converted datasets, SPARQL endpoints and documentation can be found at:

Please spread the word. We appreciate your feedback!

The LATC Team

LOD2 Webinar Series: LIMES – Discovery of Links across Knowledge Bases

March 26, 2012 - 2:03 pm by Nadine Jaenicke - 2 comments »

The 1st version of the LOD2 Stack has been published in September 2011 in the form of an LOD2 Stack demo and the downloadable LOD2 Stack virtual machine image – additional details and the instructions on installing the LOD2 Stack from scratch are available in the How-To-Start document.

Born from the wish to make linking tractable, the Link Discovery Framework for Metric Spaces (LIMES) is tailored towards the time-efficient and lossless discovery of links across knowledge bases. LIMES is an extensible declarative framework that encapsulates manifold algorithms dedicated to the processing of structured data of any sort. Built with extensibility and easy integration in mind, LIMES allows implementing applications that integrate, consume and/or generate Linked Data. Within LOD2, it will be used for discovering links between knowledge bases.

This webinar will be presented by the LOD2 Partner: University of Leipzig (ULEI), Germany.

The LOD2 webinar series is powered by the LOD2 project organised and produced by the Semantic Web Company (Austria). If you are interested in Linked (Open) Data principles and mechanisms, LOD tools & services and concrete use cases that can be realised using LOD then join us in the LOD2 webinar series! The LOD2 team is looking forward to meeting you at the webinar!!

When :
27.03. 2012, 04.00pm – 05.00pm CET
Information & Registration:

The LOD2 team is looking forward to meeting you at the webinar!!

AKSW tools prominently featured in TÜBİTAK

December 1, 2011 - 9:57 pm by Sören Auer - No comments »

Börtecin Ege wrote an article on the Semantic Web in the December issue of TÜBİTAK (The Scientific and Technological Research Council of Turkey) prominently featuring DBpedia, Relfinder, LIMES, SPARQL Benchmark and other AKSW related research projects. See Semantik Web Tübitak Bilim Teknik 12 2011.