摘要
汉语框架排歧旨在在候选框架中给句子中的目标词选择一个符合其语义场景的框架。目前研究方法存在隐层向量的计算与目标词无关、忽略了句法结构信息对框架排歧的影响等缺陷。针对上述问题,该文使用GCN对句法结构信息进行建模;引入门机制过滤隐层向量中与目标词无关的噪声信息;并在此基础上,提出一种约束机制来约束模型的学习,改进向量表示。该模型在CFN、FN1.5和FN1.7数据集上优于当前最好模型,证明了该方法的有效性。
Chinese frame disambiguation aims to select a proper frame to match the semantic scene of the target word in a specific sentence.This paper proposed a GCN based method for Chinese frame disambiguation to model the syntactic structure in the sentence.A gate mechanism is introduced to filter the noise information irrelevant to the target word in the hidden layer vector.On this basis,a constraint mechanism is proposed to optimize the learning of the model and improve the representation vector.The model outperforms the state-of-the-art models on the CFN,FN1.5 and FN1.7 datasets.
作者
游亚男
李茹
苏雪峰
闫智超
孙民帅
王超
YOU Ya'nan;LI Ru;SU Xuefeng;YAN Zhichao;SUN Minshuai;WANG Chao(School of Computer and Information Technology,Shanxi University,Taiyuan,Shanxi 030006,China;Key Laboratory of Ministry of Education for Computational Intelligence and Chinese Information Processing,Shanxi University,Taiyuan,Shanxi 030006,China;School of Modern Logistics,Shanxi Vocational University of Engineering Science and Technology,Jinzhong,Shanxi 030609,China)
出处
《中文信息学报》
CSCD
北大核心
2024年第3期33-41,共9页
Journal of Chinese Information Processing
基金
国家自然科学基金(61936012)
中新语言智能国际联合实验室(110037901001)
山西工程科技职业大学校科研基金计划项目(KJ202203)
山西省“四个一批”科技兴医创新计划项目(2022XM01)
面向战略性新兴产业的企业创新能力画像自动生成关键技术研究及应用(202102020101008)。
关键词
汉语框架排歧
句法信息
GCN
门机制
Chinese frame disambiguation
syntactic information
GCN
gate mechanism
