Text Multi-classification Based on Word Embedding and Multi-Grained Cascade Forest
Weiyu Wang, Gang He, Xinwen Liu · 2019
As an important research field in data mining and machine learning, text classification has been widely used in news classification, sentiment analysis, opinion mining, and so on. It has become an important basic subtask in the field of natural language processing. In this paper, we propose a text multi-classification algorithm based on word embedding and multi-grained cascade forest, which use word embedding model to represent text, sliding windows to scan the word vectors and cascade forest to classify texts. Experiments show that this text multi-classification algorithm can greatly improve the classification accuracy compared with other traditional machine learning models. Compared with the deep neural network classification models, it can greatly improve efficiency with achieving highly competitive performance.