公开(公告)号：US20180197530A1

公开(公告)日：2018-07-12

申请号：US15399843

申请日：2017-01-06

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Aaron K. BAUGHMAN ,  Stephen C. HAMMER ,  Christopher Eythan HOLLADAY ,  David A. PROVAN

IPC分类号： G10L15/06 ,  G10L15/183 ,  G06F17/27

CPC分类号： G10L15/063 ,  G06F17/2705 ,  G06F17/277 ,  G06F17/2785 ,  G06F17/2795 ,  G10L15/183 ,  G10L15/30 ,  G10L2015/0635

摘要： Methods, computer program products, and systems are presented. The methods include, for instance: collecting various word data from cross-domain sources and subject websites; assessing relevancy of feature vectors from external domains, live content of subject websites, and secondary terms derived from the live contents; expanding a language model for a domain by relevance passing a logistic regression threshold.

公开(公告)号：US20180197531A1

公开(公告)日：2018-07-12

申请号：US15400169

申请日：2017-01-06

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Aaron K. BAUGHMAN ,  Stephen C. HAMMER ,  Christopher Eythan HOLLADAY ,  David A. PROVAN

IPC分类号： G10L15/06 ,  G10L15/02 ,  G06F17/27 ,  G10L15/18

CPC分类号： G06F17/2735 ,  G06F17/2785 ,  G06F17/2795 ,  G10L15/183

摘要： Methods, computer program products, and systems are presented. The methods include, for instance: determining that one or more word of a feature vector more supports than negates a language model corresponding to the domain based on a sensitivity of respective word. Words having acceptable sensitivities are added to the language model, and the language model is enhanced by use of machine learning in order to accurately and comprehensively model the language specific for the domain.