The ATILF-LLF System for Parseme Shared Task: a Transition-based Verbal Multiword Expression Tagger
Hazem Al Saied, Matthieu Constant, Marie Candito · 2017
We describe the ATILF-LLF system built for the MWE 2017 Shared Task on automatic identification of verbal multiword expressions.We participated in the closed track only, for all the 18 available languages.Our system is a robust greedy transition-based system, in which MWE are identified through a MERGE transition.The system was meant to accommodate the variety of linguistic resources provided for each language, in terms of accompanying morphological and syntactic information.Using per-MWE Fscore, the system was ranked first 1 for all but two languages (Hungarian and Romanian).Transition Configuration ([ ], [1, 2, 3, 4, 5, 6], [ ]) Shift ⇒ ([1], [2, 3, 4, 5, 6], [ ]) Complete ⇒ ([ ], [2, 3, 4, 5, 6], [1]) Shift ⇒ ([2], [3, 4, 5, 6], [1]) Shift ⇒ ([2, 3], [4, 5, 6], [1]) Complete ⇒ ([2], [4, 5, 6], [1, 3]) Shift ⇒ ([2, 4], [5, 6], [1, 3]) Complete ⇒ ([2], [5, 6], [1, 3, 4]) Shift ⇒ ([2, 5], [6], [1, 3, 4]) Shift ⇒ ([2, 5, 6], [ ], [1, 3, 4]) Merge ⇒ ([2, (5, 6)], [ ], [1, 3, 4]) Merge ⇒ ([(2, (5, 6))], [ ], [1, 3, 4]) Complete ⇒ ([ ], [ ], [1, 3, 4, (2, (5, 6))])