Datenschutzerklärung|Data Privacy

Juan Soto

24.04.2015: Multilingual Semantic Parsing Paper Accepted @ ACL 2015

We are happy to announce that a paper called " Generating High Quality Proposition Banks for Multilingual Semantic Role Labeling " by Alan Akbik, Laura Chiticariu, Marina Danilevsky, Yunyao Li, Shivakumar Vaithyanathan and Huaiyu Zhu has been accepted for publication at ACL 2015 in Beijing, China . The paper has been written in collaboration with researchers from IBM's Almaden Research Center in San Jose, California.

Semantic role labeling (SRL) is crucial to natural language understanding as it identifies the predicate-argument structure in text with semantic labels. Unfortunately, resources required to construct SRL models are expensive to obtain and simply do not exist for most languages. In this paper, we present a two-stage method to enable the construction of SRL models for resource-poor languages by exploiting monolingual SRL and multilingual parallel data. Experimental results show that our method outperforms existing methods. More importantly, our method consistently produces SRL models with high to reasonable quality for 7 languages in three language families.