fasttext
链接: 数据集 提取码: 6cgu
《Bag of Tricks for Efficient Text Classification》
-高效率文本分类技巧包
作者:Armand Joulin等
单位:Facebook AI Research
发表会议及时间:EACL 2017
一 论文导读
1.文本分类简介
2.相关技术
1.文本分类简介
Input:
A document d
A fixed set of classes C={c1,c2,c3...,cn}
The input document can be considered as a sequence
Output:
A prediction class c // c 属于 C
2.相关技术
一 词表征
In traditional NLP, we regard words as discrete symbols: localist representation
Words can be represented by one-hot vectors: one 1, the rest 0s
motel=[000000010000]
hotel=[010000000000]
Vector dimension=number of words in vocabulary
二 词袋模型
三 词频与逆词频指数