Fabian Suchanek
Fabian Suchanek works on building large knowledge bases and connecting them to natural language. Their papers here focus on knowledge graphs and ontologies: how structured facts about entities can be gathered from sources such as Wikipedia, kept consistent, and linked to the many ways people phrase the same relation in text. The methods combine information extraction with existing semantic resources, using entity types and class hierarchies to organise what is extracted.
The most cited of these works is PATTY: A Taxonomy of Relational Patterns with Semantic Types from 2012. Knowledge bases name relations with abstract labels, while text expresses them through varied wording. PATTY gathers such relational phrases and arranges them into synonym groups and a hierarchy, much as WordNet does for words. It types the arguments of each pattern with semantic classes from a knowledge base, which separates phrases whose meaning depends on context. It generalises patterns both syntactically and semantically. It infers that one pattern subsumes another when the entity pairs it connects largely fall within the other's set. The next most cited is YAGO3: A Knowledge Base from Multilingual Wikipedias from 2015.
YAGO3 is also the newest of the papers here, and it turns toward multilingual and cross-lingual NLP. Most knowledge bases drawn from Wikipedia use only the English edition. That misses entities and facts that are covered only in other languages, often because they are of local interest. YAGO3 extends the YAGO knowledge base by merging several language editions into a single coherent resource, rather than building a separate one per language. Suchanek's current institution is Télécom Paris, with the Laboratoire Traitement et Communication de l’Information and the Institut Polytechnique de Paris. Earlier affiliations include the Max Planck Institute for Informatics and the Institut national de recherche en sciences et technologies du numérique. Co-authors include Gerhard Weikum.
Fabian Suchanek has 2 papers in our library, in 2012–2015, most of them on Multilingual & cross-lingual NLP, Knowledge graphs & ontologies; often with Gerhard Weikum.
current affiliationTélécom Paris, Palaiseau, France; Université Paris-Saclay, Gif-sur-Yvette, France; Institut Polytechnique de Paris, Palaiseau, France; Laboratoire Traitement et Communication de l’Information, Palaiseau, France
OpenAlex · ORCID · h-index 37 · 199 works · 12 325 citations OpenAlex, 2026-10-14
2 works in our library · 216 citations in our library · 2 more works in our library · ⭐ Paper Star rank #2547
Affiliations
- Télécom Paris · Palaiseau, France — current · 2010–2026 on 86 papers · employment from 2013 (Wikidata)
- Institut Polytechnique de Paris · Palaiseau, France — current · 2018–2026 on 48 papers
- Laboratoire Traitement et Communication de l’Information · Palaiseau, France — current · 2012–2026 on 40 papers
- Université Paris-Saclay · Gif-sur-Yvette, France — current · 2014–2026 on 2 papers
- ParisTech · Paris, France — 2013–2022 on 10 papers
- Institut national de recherche en sciences et technologies du numérique · Le Chesnay-Rocquencourt, France — 2010–2020 on 17 papers
- Max Planck Institute for Informatics · Saarbrücken, Germany — 2007–2015 on 40 papers · education until 2008 (Wikidata)
- Max Planck Society · Munich, Germany — 2006–2013 on 18 papers
- Centre Inria de Saclay · Palaiseau, France — 2010–2012 on 12 papers
- Saarland University · Saarbrücken, Germany — education until 2005 (Wikidata)
Topics
Works by year
2010s
Papers
- PATTY: A Taxonomy of Relational Patterns with Semantic Types (2012 · 432 citations)
Large knowledge bases store relations as abstract labels such as bornIn, but they do not record the many natural-language phrasings that express those relations. - YAGO3: A Knowledge Base from Multilingual Wikipedias (2015 · 235 citations)
Most Wikipedia-derived knowledge bases rely almost entirely on the English edition. This leaves out many entities and facts that appear only in other language editions, often because they are too local for English Wikipedia. YAGO3 extends YAGO by combining ten Wikipedia editions into one consistent knowledge base…
Also in our library
Co-authors here
Cited by
This page in Markdown → https://md.research.elamur.ai/free/en/authors/fabian-suchanek/