Text Classification Model Based on fastText

Most text classification models based on traditional machine learning algorithms have problems such as curse of dimensionality and poor performance. In order to solve the above problems, this paper proposes a text classification model based on fastText. Our model explores the important information contained in the text through the feature engineering, and obtains the low-dimensional, continuous and high-quality text representation through the fastText algorithm. The experiment is based on Python to classify the text dataset of "user comment data emotional polarity judgment" in Baidu Dianshi platform. In the emotional polarity judgment task, the experimental results show that the precision, recall and F values of our model are superior to the model based on traditional machine learning algorithms and have excellent classification performance.