文献类型：专利

英文题名：Probability associated feature based news event analyzing method, involves obtaining associated probability of word vectors and optimum current feature vector value for estimating situation of current news events in soft sensor model state

作者：CHENG H;YAO J;MA Y;FANG Y

机构：[1]UNIV EAST CHINA SCI & TECHNOLOGY

申请号：CN107239562-A

申请日：2017-06-13

公开日：2017-10-10

语种：英文

收录：DERWENT

摘要：NOVELTY - The method involves collecting news event information and long term characteristic lexical item information in a dictionary text. Current news text information of a current feature set characteristic lexical item is obtained. A feature set of the characteristic lexical item is obtained in an associated door. An associated probability of two word vectors are Obtained. An optimal characteristic value of the current characteristic lexical item is calculated. An optimum current feature vector value is obtained for estimating situation of current news events in soft sensor model state. USE - Probability associated feature based news event analyzing method. ADVANTAGE - The method enables improving fusion effect of the event and reliability of state estimation. DESCRIPTION OF DRAWING(S) - The drawing shows a flow diagram illustrating a Probability associated feature based news event analyzing method. '(Drawing includes non-English language text)'