• 综合
  • 标题
  • 关键词
  • 摘要
  • 学者
  • 期刊-刊名
  • 期刊-ISSN
  • 会议名称
搜索

作者:

Wang, Ming Yang (Wang, Ming Yang.) | Li, Chen Liang (Li, Chen Liang.) | Sun, Jian Dong (Sun, Jian Dong.) | Xu, Wei Ran (Xu, Wei Ran.) | Gao, Sheng (Gao, Sheng.) | Zhang, Ya Hao (Zhang, Ya Hao.) | Wang, Pu (Wang, Pu.) | Li, Jun Liang (Li, Jun Liang.)

收录:

EI Scopus

摘要:

Text comprehension and information retrieval are two essential methods which could be reinforced by modeling to semantic similarity in sentences and phrases. However, there are general problems of traditional methods on LSTM which is used to process the input sentences. Those semantic vectors cannot fully represent the entire information sequence and the information contained in firstly input content will be diluted or overwritten by the late r information. The longer the input sequence, the more serious this phenomenon is. In order to address these problems, we propose new methods with self-attention. It can incorporate weights of special words and highlight the comparison of the similarity in key words. Compared with normal self-attention which can only incorporate the weight of the key words into the naive sentences and describe position information on sentences through position encoding. Our experiment shows that new method can improve the performance of model. © 2018 IEEE.

关键词:

Semantics Digital integrated circuits Long short-term memory

作者机构:

  • [ 1 ] [Wang, Ming Yang]Beijing University of Posts and Telecommunications, China
  • [ 2 ] [Li, Chen Liang]Beijing University of Posts and Telecommunications, China
  • [ 3 ] [Sun, Jian Dong]Beijing University of Posts and Telecommunications, China
  • [ 4 ] [Xu, Wei Ran]Beijing University of Posts and Telecommunications, China
  • [ 5 ] [Gao, Sheng]Beijing University of Posts and Telecommunications, China
  • [ 6 ] [Zhang, Ya Hao]Beijing University of Technology, China
  • [ 7 ] [Wang, Pu]Beijing University of Posts and Telecommunications, China
  • [ 8 ] [Li, Jun Liang]Luoyang Electronic Equipment Test Center, China

通讯作者信息:

电子邮件地址:

查看成果更多字段

相关关键词:

相关文章:

来源 :

年份: 2018

页码: 16-19

语种: 英文

被引次数:

WoS核心集被引频次: 0

SCOPUS被引频次: 3

ESI高被引论文在榜: 0 展开所有

万方被引频次:

中文被引频次:

近30日浏览量: 0

归属院系:

在线人数/总访问数:525/4964372
地址:北京工业大学图书馆(北京市朝阳区平乐园100号 邮编:100124) 联系我们:010-67392185
版权所有:北京工业大学图书馆 站点建设与维护:北京爱琴海乐之技术有限公司