Research on the Text Clustering of Electricity Production Safety Hazards Based on K-Means Algorithm

At present, the text classification of power production safety hazards mainly relies on the traditional manual and empirical judgment. It is inefficient and cannot guarantee its correctness. To address these problems, a text feature extraction method based on Word2vec and convolutional neural networks (CNN) is proposed and combined with K-Means clustering algorithm to automatically classify power production safety hazard text. The method is based on the pre-trained Word2vec vectors and retrained based on its own text library of power production safety hazards. It combines the formal representation of low-dimensional vectors and the deep semantic features of the text, and finally clusters the text using a clustering algorithm. The experimental results show that the method can effectively improve the efficiency and accuracy of text classification.