WordGraph - Development of a unified graph-theoretical system for acquiring lexico-semantic phenomena

• Antragsteller: Professor Dr. Hinrich Schütze
• Fachliche Zuordnung: Allgemeine und Vergleichende Sprachwissenschaft, Experimentelle Linguistik, Typologie, Außereuropäische Sprachen
• Förderung: Förderung von 2007 bis 2018
• Projektkennung: Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 42840215

• Erstellungsjahr: 2018

Zusammenfassung der Projektergebnisse

We achieved good results for several of the research questions that were the subject of this proposal and published them in the top venues of our field. The most important ones are the following two: - Our inability to compute similarity efficiently for large graphs had been a big problem initially. That we were able to solve it by creating CoSimRank was an important breakthrough. - We were able to integrate traditional computational linguistics (the WordNet graph) and modern embedding technology into a common formal framework. We published several resources that will be helpful for future research on graphs and embedding in natural language processing.