首页> 外国专利> NEURAL NETWORK RELATIONSHIP EXTRACTION METHOD, COMPUTER DEVICE, AND READABLE STORAGE MEDIUM

NEURAL NETWORK RELATIONSHIP EXTRACTION METHOD, COMPUTER DEVICE, AND READABLE STORAGE MEDIUM

机译:神经网络关系提取方法,计算机设备和可读存储介质

摘要

A neural network relationship extraction method, a computer device, and a computer readable storage medium. The method comprises: constructing a dual-channel neural network model (S101); obtaining a sentence to be processed (S102); performing dependency syntactic analysis on the sentence, and obtaining two clauses of the sentence (S103); inputting the two clauses into a first channel, performing feature extraction by means of a CNN model, and obtaining first extraction information (S104); inputting the sentence into a second channel, performing feature extraction by means of a LSTM model, and obtaining second extraction information (S105); performing weighted summarization on the first extraction information and the second extraction information by means of an attention mechanism to obtain final extraction features of the sentence, and inputting the final extraction features into a softmax layer to complete the classification of relationship categories between target entities (S106). The neural network relationship extraction method can perform high-quality relationship extraction.
机译:神经网络关系提取方法,计算机设备和计算机可读存储介质。该方法包括:构建双通道神经网络模型(S101);获取要处理的句子(S102);对句子执行依赖性句法分析,并获得句子的两个子句(S103);将两个子句输入第一信道,通过CNN模型执行特征提取,并获得第一提取信息(S104);将句子输入到第二通道,通过LSTM模型执行特征提取,并获得第二提取信息(S105);通过注意机制对第一提取信息和第二提取信息执行加权摘要,以获得句子的最终提取特征,并将最终提取特征输入到Softmax层中以完成目标实体之间的关系类别的分类(S106 )。神经网络关系提取方法可以执行高质量的关系提取。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号