CC-BY Fabian M. Suchanek


Knowledge Bases

A knowledge base (KB) is a structured, computer-processable description of the world. A KB can be thought of as a graph, in which the nodes are entities and the edges are relations. Here is an example:
A knowledge base is a graph
KBs serve all kinds of purposes, such as natural language understanding in chatbots, intelligent assistance, or Web search. Our research is guided by specific real-world problems on KBs. I work together with my colleagues and students in the DIG Team of Télécom Paris to formalize problems, to design principled models for their solution, and to develop real systems that produce that solution.
See a talk See a one page summary


We are launching our project on natural language processing and knowledge bases with 4 industrial partners. We are hiring 4 PhD students and 3 engineers or postdocs.
Apart from that, we work on several projects around knowledge bases:
Logo of YAGO
Knowledge Base Construction
We work on extracting computer-readable information from Web sources. Our flagship project in this domain is YAGO, a large knowledge base constructed from Wikidata,, and other Web sources. We have also developed several approaches to query knowledge bases efficiently. One approach is based on Bash commands
Information Extraction
Our main project in this area is the NoRDF project, which extracts complex information from natural language text. We also work on extracting commercial products from the Web.
Data mining
We work on algorithms that can find automatically where a knowledge base is missing information. Our flagship project here is the AMIE rule mining system, but we also work on determining the completeness of entities, or the necessity of attributes. Here is an overview on our work of mining completeness in knowledge bases.
Recently, we have started to work on making machine learning algorithms explainable. We are also analysing the limits of deep learning.

See the list of publications for more projects.

Students, engineers, & postdocs

Deep Learning
Rule Mining
NoRDF project
OTELO project
Former students & postdocs

Further information

No free view? No review!
I no longer review for scientific journals that do not publish their articles in open access. I think that it is absurd that journals do not remunerate the authors, and do not remunerate the reviewers, but take money if someone wants to access the articles. It's like selling scientists their own work. It hinders scientific progress, especially in poorer places. Small fees that cover the processing are OK.