Research Article
Research on Chinese Patent Text Classification Based on SVM
@INPROCEEDINGS{10.4108/eai.26-5-2023.2334244, author={Ting Han}, title={Research on Chinese Patent Text Classification Based on SVM}, proceedings={Proceedings of the 2nd International Conference on Mathematical Statistics and Economic Analysis, MSEA 2023, May 26--28, 2023, Nanjing, China}, publisher={EAI}, proceedings_a={MSEA}, year={2023}, month={7}, keywords={chinese patent text classification word2vec svm machine learning}, doi={10.4108/eai.26-5-2023.2334244} }- Ting Han
Year: 2023
Research on Chinese Patent Text Classification Based on SVM
MSEA
EAI
DOI: 10.4108/eai.26-5-2023.2334244
Abstract
In recent years, the substantial increase in the number of patent applications has brought great challenges to the classification of patent texts. In order to improve the efficiency of patent text classification and further improve the level of patent management. In this paper, a machine learning method based on SVM for Chinese patent text classification model is proposed. The research uses word vectors of jieba word segmentation and word2vec model for text representation, and uses five machine learning algorithms for text classification tests. After comparing the results, the SVM model is superior to other models in accuracy, recall, F1_score, etc. This research has important guiding significance for automatic classification of patent texts.
