An Intelligent Analysis Method of Text Sentiment Tendency Based on BP Neural Network

There are more and more platforms that can be exchanged on the network. In these platforms, a variety of critical texts are also growing explosively. It is of great significance to classify these texts and tap the potential value in the texts. The tendency classification of texts is mainly by means of dictionaries and corpora. This method has a good effect in text classification, but this method relies too much on external elements such as dictionaries, and the quality of classification results is directly related to external elements such as dictionaries, and the amount of calculation is large. Therefore, this paper introduces BP neural network, uses the vector representation model of text and the mechanism of improving emotional attention, first purifies the comments of each platform, and then categorizes the comments by tendency, in many text data, the shallow level of performance is The information and value that do not come out can be mined. Based on the analysis and experiment, the accuracy of the model is basically maintained at 67.3%. If there are enough samples, the analysis efficiency is improved by 62.1%.