- 专利标题: Long-term memory networks for knowledge extraction from text and publications
-
申请号: US15370305申请日: 2016-12-06
-
公开(公告)号: US10963789B2公开(公告)日: 2021-03-30
- 发明人: Fenglong Ma , Radha Chitta , Jing Zhou , Palghat S. Ramesh , Tong Sun , Saurabh Singh Kataria
- 申请人: Xerox Corporation
- 申请人地址: US CT Norwalk
- 专利权人: Xerox Corporation
- 当前专利权人: Xerox Corporation
- 当前专利权人地址: US CT Norwalk
- 代理机构: Ortiz & Lopez, PLLC
- 代理商 Kermit D. Lopez; Luis M. Ortiz
- 主分类号: G06N5/02
- IPC分类号: G06N5/02 ; G06N5/04 ; G06F17/27 ; G06K9/62 ; G06N3/08 ; G06N3/04 ; G06N3/063
摘要:
A long-term memory network method and system for text comprehension. A recurrent neural network can be provided, which includes an external memory module and a long-short term memory unit, wherein said recurrent neural network encodes raw text information into vector representations, forms memories, finds relevant sentences to answer questions, and generates multi-word answers to said questions utilizing the long short term memory unit.
公开/授权文献
信息查询