Lisbon: Evaluating TurboSemanticParser on Multiple Languages and Out-of-Domain Data

TL;DR: The experiments have shown that, even though the parser’s performance in Chinese and Czech attains around 80% (not too far from English performance), domain shift is a serious issue, suggesting domain adaptation as an interesting avenue for future research.

Abstract: As part of the SemEval-2015 shared task on Broad-Coverage Semantic Dependency Parsing, we evaluate the performace of our last year’s system (TurboSemanticParser) on multiple languages and out-of-domain data. Our system is characterized by a feature-rich linear model, that includes scores for first and second-order dependencies (arcs, siblings, grandparents and co-parents). For decoding this second-order model, we solve a linear relaxation of that problem using alternating directions dual decomposition (AD 3 ). The experiments have shown that, even though the parser’s performance in Chinese and Czech attains around 80% (not too far from English performance), domain shift is a serious issue, suggesting domain adaptation as an interesting avenue for future research.