baidu_39633459的博客

【论文阅读】Document-Level Relation Extraction with Reconstruction1. motivationwhen a human understands a document with relationships, he or she pays more attention to learn entity pairs with relationships rather than ones without relationships.however, t

词向量one-hot representation词库V={apple, going, I, home, Machine,Learning}apple={1,0,0,0,0,0}machine={0,0,0,0,1,0}learning={0,0,0,0,0,1}其长度为词库的大小，length=|v|=10^5缺点：稀疏表示不能表达词与词相似性，句子之间相似性可以计算，但是单词之间相似性没法表示表达能力弱（如只用8维来表示单词，one-hot只能表示8个词，而词向量可以表示2^8或者