Open AccessResearch on Application of Emergency Text Classification Based on Combination Model
Author(s) -
Ling Long,
Yafei Song,
Dan Liu
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/719/1/012079
Subject(s) - softmax function , computer science , convolutional neural network , feature (linguistics) , artificial intelligence , recall , layer (electronics) , deep learning , event (particle physics) , process (computing) , natural language processing , pattern recognition (psychology) , data mining , machine learning , linguistics , philosophy , chemistry , physics , organic chemistry , quantum mechanics , operating system
Based on the powerful learning ability of the deep learning model, the text uses convolutional memory network (CLSTM) to process the text information of the emergency. Firstly, the convolutional neural network is used to learn the local spatial feature information of the text, and then the text is extracted by the long and short memory neural network. The time feature information is used to scale the feature information by using the Softmax layer to finally obtain the event text category. Through experimental comparison, the precision, recall and comprehensive values of CLSTM model reached 0.879, 0.877 and 0.848, respectively, which were significantly higher than MLP and CNN models.