关键词: 文本分类, fastText, 线性分类器, 快速文本分类模型

Abstract: A goal of text classification is to shorten the classification time under the premise of ensuring the accuracy of classification. In order to solve the problem that the current text classification process is complex and time-consuming, fastText, a library for efficient learning of word representations and sentence classification proposed by Facebook is applied to Chinese text classification field to test its effect. Compared with the mainstream text classification method, fastText can guarantee the classification accuracy and greatly shorten the classification time. At the same time, this paper further studies and analyzes the classification accuracy and parameters and gets some optimization rules.

Key words: text classification, fastText, linear classifier, fast text classification model