Application of Convolutional Neural Networks to the Classification of Agricultural Technology Articles

In recent years, China has built many popular websites for agricultural science and technology articles, and in order to solve the time-consuming and labor-intensive problem of classifying articles in such websites, this paper implements the article classification system of textCNN convolutional neural network based on Pytorch framework. Python crawler technology is used to crawl the agricultural science and technology articles of China Agriculture Network, and calibrate them according to the original classification information, and divide them into training dataset and test dataset according to the ratio of 2/8. On the training obtained model, the best effect of the test set classification is 93.33%, and this model can be used to assist relevant technical personnel to achieve rapid sorting and classification of agricultural scientific and technical articles, which has a positive effect on the rapid dissemination of agricultural information.