Semantic Role Labeling For Russian Language Based on Ensemble Model
Xinping Zheng, Bin Zhou, Jiuming Huang, Yunxuan Liu, Hao Henry Wang, Zhichao Wang · 2019
Traditional Russian semantic role labeling (SRL) methods heavily rely on complicated and hand-crafted rules or semantic dictionaries, which increase the difficulty of this task. In this paper, bidirectional Gated Recurrent Unit (Bi-GRU) and attention mechanism are respectively used to extract the potential features of arguments which are prepared to be classified, and then we input above features and the basic features of the argument into a neural network to identify its role. In addition, we use voting ensemble mechanism to determine the final category. Experiments show that our model achieves macro F1 = 78.8 and micro F1 = 81.7 on FrameBank dataset, which outperforms the previous state-of-the-art results by Δmicro F1 = 1.7, and Δmacro F1 = 2.0.