[python] LDA处理文档主题分布代码入门笔记

以前只知道LDA是个好东西，但自己并没有真正去使用过。同时，关于它的文章也非常之多，推荐大家阅读书籍《LDA漫游指南》，最近自己在学习文档主题分布和实体对齐中也尝试使用LDA进行简单的实验。这篇文章主要是讲述Python下LDA的基础用法，希望对大家有所帮助。如果文章中有错误或不足之处，还请海涵~

一. 下载安装

LDA推荐下载地址包括：其中前三个比较常用。

gensim下载地址：https://radimrehurek.com/gensim/models/ldamodel.html

pip install lda安装地址：https://github.com/ariddell/lda

scikit-learn官网文档：LatentDirichletAllocation

其中sklearn的代码例子可参考下面这篇：

Topic
extraction with NMF and Latent Dirichlet Allocation

其部分输出如下所示，包括各个主体Topic包含的主题词：

[plain] view
plain copy

Loading dataset...

Fitting LDA models with tf features, n_samples=2000 and n_features=1000...

done in 0.733s.

Topics in LDA model:

Topic #0:

000 war list people sure civil lot wonder say religion america accepted punishment bobby add liberty person kill concept wrong

Topic #1:

just reliable gods consider required didn war makes little seen faith default various civil motto sense currency knowledge belief god

Topic #2:

god omnipotence power mean rules omnipotent deletion policy non nature suppose definition given able goal nation add place powerful leaders

....

下面这三个也不错，大家有时间的可以见到看看：

https://github.com/arongdari/python-topic-model

https://github.com/shuyo/iir/tree/master/lda

https://github.com/a55509432/python-LDA

其中第三个作者a55509432的我也尝试用过，模型输出文件为：

model_parameter.dat 保存模型训练时选择的参数
wordidmap.dat 保存词与id的对应关系，主要用作topN时查询
model_twords.dat 输出每个类高频词topN个
model_tassgin.dat 输出文章中每个词分派的结果，文本格式为词id:类id
model_theta.dat 输出文章与类的分布概率，文本一行表示一篇文章，概率1 概率2..表示文章属于类的概率
model_phi.dat 输出词与类的分布概率，是一个K*M的矩阵，K为设置分类的个数，M为所有文章的词的总数

但是短文本信息还行，但使用大量文本内容时，输出文章与类分布概率几乎每行数据存在大量相同的，可能代码还存在BUG。

下面是介绍使用pip install lda安装过程及代码应用：

[python] view
plain copy

pip install lda



参考：[python] 安装numpy+scipy+matlotlib+scikit-learn及问题解决

二. 官方文档

这部分内容主要参考下面几个链接，强推大家去阅读与学习：

官网文档：https://github.com/ariddell/lda

lda: Topic modeling with latent Dirichlet Allocation

Getting started with Latent Dirichlet Allocation in Python - sandbox

[翻译] 在Python中使用LDA处理文本 - letiantian

文本分析之TFIDF/LDA/Word2vec实践 - vs412237401

1.载入数据

[python] view
plain copy

import numpy as np

import lda

import lda.datasets

# document-term matrix

X = lda.datasets.load_reuters()

print("type(X): {}".format(type(X)))

print("shape: {}\n".format(X.shape))

print(X[:5, :5])

# the vocab

vocab = lda.datasets.load_reuters_vocab()

print("type(vocab): {}".format(type(vocab)))

print("len(vocab): {}\n".format(len(vocab)))

print(vocab[:5])

# titles for each story

titles = lda.datasets.load_reuters_titles()

print("type(titles): {}".format(type(titles)))

print("len(titles): {}\n".format(len(titles)))

print(titles[:5])

载入LDA包数据集后，输出如下所示：

X矩阵为395*4258，共395个文档，4258个单词，主要用于计算每行文档单词出现的次数（词频），然后输出X[5,5]矩阵；

vocab为具体的单词，共4258个，它对应X的一行数据，其中输出的前5个单词，X中第0列对应church，其值为词频；

titles为载入的文章标题，共395篇文章，同时输出0~4篇文章标题如下。

[python] view
plain copy

type(X): <type 'numpy.ndarray'>

shape: (395L, 4258L)

[[ 1 0 1 0 0]

[ 7 0 2 0 0]

[ 0 0 0 1 10]

[ 6 0 1 0 0]

[ 0 0 0 2 14]]

type(vocab): <type 'tuple'>

len(vocab): 4258

('church', 'pope', 'years', 'people', 'mother')

type(titles): <type 'tuple'>

len(titles): 395

('0 UK: Prince Charles spearheads British royal revolution. LONDON 1996-08-20',

'1 GERMANY: Historic Dresden church rising from WW2 ashes. DRESDEN, Germany 1996-08-21',

"2 INDIA: Mother Teresa's condition said still unstable. CALCUTTA 1996-08-23",

'3 UK: Palace warns British weekly over Charles pictures. LONDON 1996-08-25',

'4 INDIA: Mother Teresa, slightly stronger, blesses nuns. CALCUTTA 1996-08-25')

From the above we can see that there are 395 news items (documents) and a vocabulary of size 4258. The document-term matrix, X, has a count of the number of occurences of each of the 4258 vocabulary
words for each of the 395 documents.

下面是测试文档编号为0，单词编号为3117的数据，X[0,3117]：

[python] view
plain copy

# X[0,3117] is the number of times that word 3117 occurs in document 0

doc_id = 0

word_id = 3117

print("doc id: {} word id: {}".format(doc_id, word_id))

print("-- count: {}".format(X[doc_id, word_id]))

print("-- word : {}".format(vocab[word_id]))

print("-- doc : {}".format(titles[doc_id]))

'''''输出

doc id: 0 word id: 3117

-- count: 2

-- word : heir-to-the-throne

-- doc : 0 UK: Prince Charles spearheads British royal revolution. LONDON 1996-08-20

'''

2.训练模型

其中设置20个主题，500次迭代

[python] view
plain copy

model = lda.LDA(n_topics=20, n_iter=500, random_state=1)

model.fit(X) # model.fit_transform(X) is also available

3.主题-单词（topic-word）分布

代码如下所示，计算'church', 'pope', 'years'这三个单词在各个主题(n_topocs=20，共20个主题)中的比重，同时输出前5个主题的比重和，其值均为1。

[python] view
plain copy

topic_word = model.topic_word_

print("type(topic_word): {}".format(type(topic_word)))

print("shape: {}".format(topic_word.shape))

print(vocab[:3])

print(topic_word[:, :3])

for n in range(5):

sum_pr = sum(topic_word[n,:])

print("topic: {} sum: {}".format(n, sum_pr))

输出结果如下：

[python] view
plain copy

type(topic_word): <type 'numpy.ndarray'>

shape: (20L, 4258L)

('church', 'pope', 'years')

[[ 2.72436509e-06 2.72436509e-06 2.72708945e-03]

[ 2.29518860e-02 1.08771556e-06 7.83263973e-03]

[ 3.97404221e-03 4.96135108e-06 2.98177200e-03]

[ 3.27374625e-03 2.72585033e-06 2.72585033e-06]

[ 8.26262882e-03 8.56893407e-02 1.61980569e-06]

[ 1.30107788e-02 2.95632328e-06 2.95632328e-06]

[ 2.80145003e-06 2.80145003e-06 2.80145003e-06]

[ 2.42858077e-02 4.66944966e-06 4.66944966e-06]

[ 6.84655429e-03 1.90129250e-06 6.84655429e-03]

[ 3.48361655e-06 3.48361655e-06 3.48361655e-06]

[ 2.98781661e-03 3.31611166e-06 3.31611166e-06]

[ 4.27062069e-06 4.27062069e-06 4.27062069e-06]

[ 1.50994982e-02 1.64107142e-06 1.64107142e-06]

[ 7.73480150e-07 7.73480150e-07 1.70946848e-02]

[ 2.82280146e-06 2.82280146e-06 2.82280146e-06]

[ 5.15309856e-06 5.15309856e-06 4.64294180e-03]

[ 3.41695768e-06 3.41695768e-06 3.41695768e-06]

[ 3.90980357e-02 1.70316633e-03 4.42279319e-03]

[ 2.39373034e-06 2.39373034e-06 2.39373034e-06]

[ 3.32493234e-06 3.32493234e-06 3.32493234e-06]]

topic: 0 sum: 1.0

topic: 1 sum: 1.0

topic: 2 sum: 1.0

topic: 3 sum: 1.0

topic: 4 sum: 1.0

4.计算各主题Top-N个单词

下面这部分代码是计算每个主题中的前5个单词

[python] view
plain copy

n = 5

for i, topic_dist in enumerate(topic_word):

topic_words = np.array(vocab)[np.argsort(topic_dist)][:-(n+1):-1]

print('*Topic {}\n- {}'.format(i, ' '.join(topic_words)))

输出如下所示：

[python] view
plain copy

*Topic 0

- government british minister west group

*Topic 1

- church first during people political

*Topic 2

- elvis king wright fans presley

*Topic 3

- yeltsin russian russia president kremlin

*Topic 4

- pope vatican paul surgery pontiff

*Topic 5

- family police miami versace cunanan

*Topic 6

- south simpson born york white

*Topic 7

- order church mother successor since

*Topic 8

- charles prince diana royal queen

*Topic 9

- film france french against actor

*Topic 10

- germany german war nazi christian

*Topic 11

- east prize peace timor quebec

*Topic 12

- n't told life people church

*Topic 13

- years world time year last

*Topic 14

- mother teresa heart charity calcutta

*Topic 15

- city salonika exhibition buddhist byzantine

*Topic 16

- music first people tour including

*Topic 17

- church catholic bernardin cardinal bishop

*Topic 18

- harriman clinton u.s churchill paris

*Topic 19

- century art million museum city

5.文档-主题（Document-Topic）分布

计算输入前10篇文章最可能的Topic

[python] view
plain copy

doc_topic = model.doc_topic_

print("type(doc_topic): {}".format(type(doc_topic)))

print("shape: {}".format(doc_topic.shape))

for n in range(10):

topic_most_pr = doc_topic
.argmax()

print("doc: {} topic: {}".format(n, topic_most_pr))

输出如下所示：

[python] view
plain copy

type(doc_topic): <type 'numpy.ndarray'>

shape: (395L, 20L)

doc: 0 topic: 8

doc: 1 topic: 1

doc: 2 topic: 14

doc: 3 topic: 8

doc: 4 topic: 14

doc: 5 topic: 14

doc: 6 topic: 14

doc: 7 topic: 14

doc: 8 topic: 14

doc: 9 topic: 8

6.两种作图分析

详见英文原文，包括计算各个主题中单词权重分布的情况：

[python] view
plain copy

import matplotlib.pyplot as plt

f, ax= plt.subplots(5, 1, figsize=(8, 6), sharex=True)

for i, k in enumerate([0, 5, 9, 14, 19]):

ax[i].stem(topic_word[k,:], linefmt='b-',

markerfmt='bo', basefmt='w-')

ax[i].set_xlim(-50,4350)

ax[i].set_ylim(0, 0.08)

ax[i].set_ylabel("Prob")

ax[i].set_title("topic {}".format(k))

ax[4].set_xlabel("word")

plt.tight_layout()

plt.show()

输出如下图所示：



第二种作图是计算文档具体分布在那个主题，代码如下所示：

[python] view
plain copy

import matplotlib.pyplot as plt

f, ax= plt.subplots(5, 1, figsize=(8, 6), sharex=True)

for i, k in enumerate([1, 3, 4, 8, 9]):

ax[i].stem(doc_topic[k,:], linefmt='r-',

markerfmt='ro', basefmt='w-')

ax[i].set_xlim(-1, 21)

ax[i].set_ylim(0, 1)

ax[i].set_ylabel("Prob")

ax[i].set_title("Document {}".format(k))

ax[4].set_xlabel("Topic")

plt.tight_layout()

plt.show()

输出结果如下图：

三. 总结

这篇文章主要是对Python下LDA用法的入门介绍，下一篇文章将结合具体的txt文本内容进行分词处理、文档主题分布计算等。其中也会涉及Python计算词频tf和tfidf的方法。

由于使用fit()总报错“TypeError: Cannot cast array data from dtype('float64') to dtype('int64') according to the rule 'safe'”，后使用sklearn中计算词频TF方法：

http://scikit-learn.org/stable/modules/feature_extraction.html#text-feature-extraction

总之，希望文章对你有所帮助吧！尤其是刚刚接触机器学习、Sklearn、LDA的同学，毕竟我自己其实也只是一个门外汉，没有系统的学习过机器学习相关的内容，所以也非常理解那种不知道如何使用一种算法的过程，毕竟自己就是嘛，而当你熟练使用后才会觉得它非常简单，所以入门也是这篇文章的宗旨吧！

最后非常感谢上面提到的文章链接作者，感谢他们的分享。如果有不足之处，还请海涵~

(By:Eastmount 2016-03-17 深夜3点半 http://blog.csdn.net/eastmount/ )