Research on Chinese Word Separation Based on Deep Learning

摘要

Currently, a large amount of information is generated every day, and natural language processing techniques can help people to get the information they need quickly. For natural language processing of Chinese, Chinese word separation is a fundamental task in natural language processing. At present, research on Chinese word separation is basically based on machine learning methods, with the disadvantage that a large number of features need to be constructed manually. To address the shortcomings of current Chinese word sorting, this paper first analyzes the common methods and deep learning models for Chinese word sorting, and proposes an improvement scheme based on the Chinese word sorting model Bi LSTM+textbf CRF. And experiments are designed to verify the correctness and superiority of the model proposed in the paper on Chinese word separation on three datasets.