We describe our CoNLL 2009 Shared Task system in the present paper. The system includes three cascaded components: a generative dependency parser, a classifier for syntactic dependency labels and a semantic classifier. The experimental results show that the labeled macro F1 scores of our system on the joint task range from 43.50% (Chinese) to 57.95% (Czech), with an average of 51.07%.

In this article, we have studied the possibility of applying Borda and Fuzzy Borda voting schemes to combine semantic role labeling systems. To better select the correct semantic role, among those provided by different experts, we have introduced two measures: the first one calculates the overlap between labeled sentences, whereas the second one adds different scoring levels depending on the ve...

The research led by Gali (AER 1999) and Basu et al. (AER 2006) raises two important questions regarding the validity of the RBC theory: (i) How important are technology shocks in explaining the business cycle? (ii) Do impulse responses to technology shocks found in the data reject the assumption of ‡exible prices? This paper argues that the conditional impulse responses of the U.S. economy to t...

SEMANTIC ROLE LABELING OF IMPLICIT ARGUMENTS FOR NOMINAL PREDICATES

In spite of the large variety of technological means available today to obtain and share knowledge needed by the progress of our Society, the relation between the knowledge produced and the knowledge required is not sufficiently facilitated. This paper aims to provide a partial answer to this issue. It proposes a NIAM/ORM formalization of exchanges between producers and users of knowledge. The ...

BACKGROUND The use of different forms of coordination according to situational demands plays a crucial role in teams working in complex environments. This study aimed to describe patterns of coordinative actions (CAs) as they occur during anaesthesia induction and to analyse the influence of two crucial situational factors on these patterns, namely the amount of existing standards and the level...

who offered both ever-insightful comments and encouragement. My committee member Matthew Dryer, who spent much of his time reading and commenting on this project, in spite of his extremely busy schedule. My committee member Leonard Talmy, who gave me support and warm encouragement.

This paper focuses on semantic role labeling using automatically-generated syntactic information. A simple and robust strategy for system combination is presented, which allows to partially recover from input parsing errors and to significantly boost results of individual systems. This combination scheme is also very flexible since the individual systems are not required to provide any informat...