首页>
外国专利>
ENTERPRISE RELATIONSHIP EXTRACTION METHOD AND DEVICE, AND STORAGE MEDIUM
ENTERPRISE RELATIONSHIP EXTRACTION METHOD AND DEVICE, AND STORAGE MEDIUM
展开▼
机译:企业关系提取方法和设备以及存储介质
展开▼
页面导航
摘要
著录项
相似文献
摘要
Disclosed are an enterprise relationship extraction method and device, and a storage medium. The method comprises: extracting sentences comprising pairs of related enterprise entities from a repository as training sample sentences to establish a sample database; extracting all training sample sentences, each of which comprises a pair of enterprise entities, and performing word segmentation, mapping each word to a word vector xi, and mapping each sentence to a sentence vector Si; using LSTM to calculate a first hidden-layer-state vector hi and a second hidden-layer-state vector hi' of the word vector xi, performing splicing to obtain a comprehensive hidden-layer-state vector, and then obtaining a feature vector Ti; substituting the feature vector Ti into an average vector expression to calculate an average vector S; substituting the average vector S and a relation type of the pair of enterprise entities into a softmax classification function to calculate a weight ai of each training sample sentence; and extracting each sentence comprising two enterprise entities, obtaining a feature vector Ti by means of a bi-LSTM, and inputting the vector into a trained RNN model to predict the relation between the two enterprises, so that labor costs are reduced, and the relation between two enterprise entities can be predicted more accurately.
展开▼
机译:公开了一种企业关系提取方法,装置和存储介质。该方法包括:从存储库中提取包括成对的相关企业实体对的句子作为训练样本句子,以建立样本数据库;以及提取所有训练样本句子,每个句子包含一对企业实体,并进行词分割,将每个词映射到词向量x i Sub>,并将每个句子映射到句子向量S 我 Sub>;使用LSTM计算单词向量x i <的第一个隐藏层状态向量h i Sub>和第二个隐藏层状态向量h i Sub>' / Sub>,进行拼接得到综合的隐层状态向量,然后得到特征向量T i Sub>;将特征向量T i Sub>代入平均向量表达式,计算出平均向量S;将平均向量S和一对企业实体的关系类型代入softmax分类函数,计算出每个训练样本句子的权重a i Sub>;提取包含两个企业实体的每个句子,通过bi-LSTM获得特征向量T i Sub>,并将该向量输入到训练有素的RNN模型中以预测两个企业之间的关系,减少了人工成本,可以更准确地预测两个企业实体之间的关系。
展开▼