pytreebank︱情感分析可视化——情感结构树

效果如上，一句话用树状来表达，具体案例可查询stanford Sentiment Analysis网站

如何用py来画，主要参考自pytreebank-github

1 直接上案例，官网案例数据下载:

http://nlp.stanford.edu/sentiment/trainDevTestTrees_PTB.zip

import pytreebank
dataset = pytreebank.load_sst("/path/to/sentiment/")
pytreebank.LabeledTree.inject_visualization_javascript()
# select and example to visualize
example = dataset["train"][0]
# display it in the page
example.display()

load_sst时候需要写入官网下载数据的链接具体在哪。

2 那么案例里面的数据长什么样子：

for label, sentence in example.to_labeled_lines():
print("%s has sentiment label %s" % (
sentence,
["very negative", "negative", "neutral", "positive", "very positive"][label]))

具体如下：

The Rock is destined to be the 21st Century 's new `` Conan '' and that he 's going to make a splash even greater than Arnold Schwarzenegger , Jean-Claud Van Damme or Steven Segal . has sentiment label positive
The Rock has sentiment label neutral
The has sentiment label neutral
Rock has sentiment label neutral
is destined to be the 21st Century 's new `` Conan '' and that he 's going to make a splash even greater than Arnold Schwarzenegger , Jean-Claud Van Damme or Steven Segal . has sentiment label very positive
is destined to be the 21st Century 's new `` Conan '' and that he 's going to make a splash even greater than Arnold Schwarzenegger , Jean-Claud Van Damme or Steven Segal has sentiment label positive
is has sentiment label neutral
destined to be the 21st Century 's new `` Conan '' and that he 's going to make a splash even greater than Arnold Schwarzenegger , Jean-Claud Van Damme or Steven Segal has sentiment label very positive
destined has sentiment label neutral
to be the 21st Century 's new `` Conan '' and that he 's going to make a splash even greater than Arnold Schwarzenegger , Jean-Claud Van Damme or Steven Segal has sentiment label neutral
to be the 21st Century 's new `` Conan '' and has sentiment label neutral
to be the 21st Century 's new `` Conan '' has sentiment label neutral
to be the 21st Century 's new `` Conan has sentiment label neutral
to has sentiment label neutral

3 那么自己咋画呢：

pytreebank.LabeledTree.inject_visualization_javascript()

pytreebank.create_tree_from_string('(4 (0 你) (3 (2 是) (3 (3 (3 谁) (2 的)) (2 谁))))').display()

4代表总共有四个节点，

(0 你)代表，‘你’这个字的情感点；

编号的意义为：

0–>”very negative”,

1–> “negative”

2–>, “neutral”

3–>, “positive”

4–>, “very positive”

4 其他

pytreebank.create_tree_from_string('(4 (0 你) (3 (2 是) (3 (3 (3 谁) (2 的)) (2 谁))))').to_dict()

把结果导出成dict形式

pytreebank.create_tree_from_string('(4 (0 你) (3 (2 是) (3 (3 (3 谁) (2 的)) (2 谁))))').to_json()

把结果导出成json形式

pytreebank.create_tree_from_string('(4 (0 你) (3 (2 是) (3 (3 (3 谁) (2 的)) (2 谁))))').to_lines()

把结果导出成一列文字形式

pytreebank.create_tree_from_string('(4 (0 你) (3 (2 是) (3 (3 (3 谁) (2 的)) (2 谁))))').to_lines()

把结果导出成to_labeled_lines形式