TextRank, 关键词和句子抽取

1. 简介

TextRank, 基于图模型的关键词和句子抽取. 与 google 的PageRank有一定的相通之处. 都是 unsupervised.

2. theory

术语约定:

vote

G=(V,E). 点 A links to B , 可以认为是 A 对 B 发起的一次vote, 即 A casts a vote for B.

TextRank model 的基本思想是:

点A得到的投票数越多, A越重要.

点A的重要性直接影响到它所投票的重要性.

2.1 formula

S(Vi)=(1−d)+d∗∑j∈In(Vi)1|Out(Vj)|S(Vj)(1)

S(Vi)

score of Vi

d

damping factor that can be set [0,1], usually set to 0.85.

In(Vi)

the set of vertices that point to Vi

Out(Vi)

the set of vertices that Vi point to .

3.build graph

基于图的排序模型, 那么关键就是建图.

3.1 edge

two vertices are connected if they co-occur within a window of maximum N words, where N can be set anywhere between [2,10].

directed edge

若要有序, the direction was set following the natural flow of the text.

undirected edge

无向边可以认为是双向的弧.

3.2 vertex

句子由若干个word组成 , 需要对这些 word 作简单的处理, 才能当作 vertices , added to the graph.

pre-processing

根据stop word, 词性标注过滤一部分, 剩下的当作vertex.

论文实验显示, 仅考虑 noun 和 adjective 是好的.

processing

根据文本构建 graph 的过程.

post-processing

相邻接的 key word sequence 会 collapsed into multi-word keyword.

论文中是这么举例的. 在文章

Matlab code for plotting ambiguity functions

中, 如果

Matlab

与

code

总是邻接, 那么就可以考虑合成为

Matlab code

, 丰富意思的表达.

4. discussion

论文中试验用的 data set is a collection of 500 abstracts from the

Inspec

database.

结果: precision:31.2, Recall:43.1, F-measure:36.2 .

window size

论文中用了(2,3,510), 取2时效果最好.

directed or undirected edge

论文中用 无向边 的时候, 效果最好.

5. sentence extraction

可以用于 自动生成摘要.

顶点

不再是单词, 而是整个句子.

边

不再以 co-occurrence 作为连接的条件, 因为 “co-occurrence” is not a meaningful relation for such large context.

而是依据 similarity.

记 句子 S=w1,w2,...,wn, 表示为单词的序列, 那么句子之间的相似度为:

Similarity(Si,Sj)=|{wk|wk∈Si∩wk∈Sj}|log(|Si|)+log(|Sj|)(5-1)

把相似度作为权重, 计算顶点的重要性:

WS(Vi)=(1−d)+d∗∑j∈In(Vi)wj,i∑k∈Out(Vj)wj,kWS(Vj)(5-2)

参考

原始论文, TextRank: Bringing Order into Texts