Automatic classification method of coal mine safety hidden danger informatio

Abstract

Manual classification method is difficult to meet classification requirements of massive coal mine safety hidden danger information, and automatic text classification method based on probability statistics has low classification accuracy rate. In view of the above problems, an automatic classification method of coal mine safety hidden danger information was proposed which was based on Word2vec and convolutional neural network. Firstly, hidden danger information is pre-processed through word segmentation and stop word deletion. Then semantic similarity between words is represented by employing Word2vec. Finally, local context high-level features of hidden danger information are extracted by use of convolutional neural network, and Softmax classifier is used to realize automatic classification of hidden danger information. The experimental results show that the method realizes end-to-end automatic classification and can effectively improve accuracy and comprehensiveness of classification.

Keywords

• coal mine safety
• automatic classification of hidden danger information
• text classification
• convolutional neural network
• word2vec