Research on Text Classification Method Based on LSTM Neural Network Model
Yanbo Zhang
Abstract
Text classification is a process of automatically classifying test data according to given rules. Word embedding technology is based on neural probabilistic language model, which can get word vectors with rich semantic information. In the task of natural language processing, a set of excellent word vectors is the basis of all researches. In order to search and extract information from massive electronic texts, this paper constructs a LSTM neural network classification model to classify text information. LSTM can extract words and sentences with different contributions, and combine LSTM's region embedding technology to classify text. Experimental results show that, compared with traditional methods, this method has obvious improvement in performance and classification accuracy.