| 摘 要: 自然语言处理是人工智能领域中的一个热门方向，而文本分类作为自然语言处理中的关键技术受到专家 学者的广泛关注。随着计算机网络的发展，海量的文本踊跃出来。文本越来越多，通过人工对文本进行分类的成本越来 越高。本文针对短文本分类问题，使用词袋模型从词向量中提取词频矩阵，删除停止词与低频词。再使用TF-IDF算法 提取文本特征，进行文本分类研究，最终可以使短文本以较高的正确率归类。 |
| 关键词: 自然语言处理；短文本分类；词袋模型；TF-IDF |
|
中图分类号: TP391.1
文献标识码: A
|
| 基金项目: 国家自然科学基金项目(61202458/61403109)；黑龙江省自然科学基金项目(F2017021). |
|
| Research on Short Text Classification Based on Bag of Words and TF-IDF |
|
HUANG Chunmei,WANG Songlei
|
( College of Computer Science and Information Engineering, Harbin Normal University, Harbin 150025, China)
|
| Abstract: Natural language processing (NLP) is a hot field in the field of artificial intelligence.Text categorization,as a key technology in NLP,has attracted extensive attention from experts and scholars.With the development of computer networks,massive texts have come out enthusiastically.As there are more and more texts,it becomes more and more expensive to classify them manually.In this paper,we use the bag of words model to extract the word frequency matrix from the word vectors and delete the stop words and low-frequency words.Then TF-IDF algorithm is used to extract text features and conduct text classification research so that the short text can be classified with high accuracy. |
| Keywords: natural language processing;short text classification;bag of words;TF-IDF |
