A New Method to Improve the Accuracy of the Chinese Language Text Classification Based on Big Data Text

The role of computers in modern society is very important. Natural Language Processing (NLP) is an important direction in the field of computer science and artificial intelligence. Most of the modern NLP fields are dominated by European and American countries. In addition, our Chinese language is very different from the phonetic characters in Europe and America. Although the development of Chinese NLP is in the ascendant, there are still many problems. As The Belt and Road Initiative is widely accepted and welcomed in the world, political, economic, cultural and other fields are accompanied by its deepening exchanges and development in the countries along the way. This paper chooses the topic of overseas Chinese text classification to carry out research, hoping to share its modest power for the development of The Belt and Road Initiative. We designed three experiments: THUCNEWS text classification based on THUCTC, the Chinese Language (CL) text classification directly using THUCTC, the CL text classification using THUCTC with CL wordlist. Although it costs us a lot of energy to build a CL wordlist, it is effective to improve the accuracy of CL text classification by 1.8%.