The Data Science (DICE) group develops methods, algorithms and applications for the extraction, integration, storage, querying, access and consumption of large-scale datasets. In contrast to most other groups, DICE focusses on knowledge-driven methods. We hence rely and extend on knowledge representation standards developed for the Semantic Web. The area of application of our research include but are not limited to solutions for federated queries on the Web, knowledge extraction from text and other types of datasets, knowledge integration and fusion, keyword-based search and question answering.


We are dedicated to open-source software and open publications. Have a look at our tool page to find a list of the open-source frameworks we offer. These tools and frameworks implement our innovative approaches to the problems aforementioned and are designed to facilitate their swift integration into industry projects. Our project page gives you an overview of the projects we have worked or are working on. Interested in working with us? Please click here to contact us.

Funded by


Source code: https://github.com/dice-group/LIMES

The provision of links between knowledge bases is one of the core principles of Linked Data. With the growth of the number and the size of RDF...

Read more

The LIMES development team is happy to announce LIMES 1.3.0!

LIMES is a link discovery framework for the Web of Data, which implements time-efficient approaches for large-scale link discovery based...

Read more

We are happy to announce that our paper, “Characterizing Mention Mismatching Problems for Improving Recognition Results”, was accepted at the 19th International Conference on Information Integration...

Read more

On March 20th, we were invited to a workshop at the TIB Hannover to discuss the vision for an Open Research Knowledge Graph (ORKG).

In the following, we will depict how the DICE research group can...

Read more

With the ever growing number and size of knowledge bases comes the necessity to find links between these data sources as well. To tackle this problem we present DRAGON, a time-efficient and accurate...

Read more

Linked Data enrichment is the process of adding, altering or deleting the set of triples of an input dataset in order to obtain an enriched version of it. This enriched dataset usually provides...

Read more


Prof. Dr. Axel-Cyrille Ngonga Ngomo
head professor
phone: +49 5251 60-3342
fax: +49 5251 60-3436
e-mail: axel.ngonga(at)upb.de
office: O4.213
Simone Auinger
phone: +49 5251 60-1764
fax: +49 5251 60-3436
e-mail: mone(at)upb.de
office: O 4.113