SFB 991 / CRC 991
SToRE member since 30.05.2017
1. advisor: Laura Kallmeyer
2. advisor: Wiebke Petersen
3. advisor: Katalin Balogh
In my dissertation project I develop a frame-semantic parser for French based on the framework of Tree-Adjoining-Grammars (TAG) and Neural TAG supertagging. Semantic parsing is a task of computational processing of natural language utterances to structured and logically interpretable meaning representations. These meaning representations can be used in downstream applications for further tasks like query execution, logical inference or reasoning. Frame-semantic parsing uses semantic frames for representation of the sentential meaning within the framework of frame semantics. The core idea of frame semantics is based on the assumption that the meaning of a single word can be defined through the essential background knowledge related to this word. During the last two decades, several frame-annotated linguistic resources were created, which initiated an increased research into frame-based semantic role labeling (SRL) and frame-semantic parsing. In the planned PhD dissertation I intend to use the available linguistic resources for French and implement a data-driven frame-semantic parsing system on top of them.
My dissertation is part of the TreeGraSP (https://treegrasp.phil.hhu.de/) project.
Tatiana Bladier, Andreas van Cranenburgh, Younes Samih, Laura Kallmeyer (2018): German and French Neural Supertagging Experiments for LTAG Parsing. ACL 2018 Student Research Workshop, Melbourne, Australia.
Tatiana Bladier, Esther Seyffarth, Oliver Hellwig, Wiebke Petersen (2018): AET: Web-based Adjective Exploration Tool for German. LREC 2018, Miyazaki, Japan.
Tatiana Bladier (2018): Frame-semantic parsing for French with LTAG supertagging. NASSLLI 2018 Student Session, Pittsburgh, USA.
Tatiana Bladier (2018): Bi-LSTM TAG Supertagging for French and German. TreeGraSP Workshop, Düsseldorf, Germany.
Tatiana Bladier, Kilian Evang, Yulia Zinova (2018): Representing Slavic Aktionsarten in DRT with Boxer. DGfS 2018, Stuttgart, Germany.
Tatiana Bladier, Kilian Evang, Yulia Zinova (2017): Aspecto-temporal Representation of Slavic Aktionsarten in DRT with Boxer. LENLS-14, Tokyo, Japan.