[PYTHON] Work log when scraping and applying LDA

Overview

I scraped from some URLs in Ruby and extracted topics using LDA in Python.

1. Scraping
2. Morphological analysis
3. Original dictionary
4. Data shaping
5. LDA execution

1. Scraping

Use of Mechanize

gem install

$ bundle init
$ vim Gemfile

gem 'mechanize'

$ bundle install

Use of mechanize

It is OK if it works with the following sample file.

sample.rb

require 'mechanize'

agent = Mechanize.new
search_page = agent.get('Appropriate URL')

search_page.search('body p').each do |y|
  p y.text
end

2. Morphological analysis

Install Mecab on Mac

$ brew search mecab
mecab mecab-ipadic

$ brew install mecab mecab-ipadic

$ mecab

OK if mecab starts

Use of Natto

natto is a gem that wraps mecab installed on your system.

gem install

$ bundle init
$ vim Gemfile

gem 'natto'

$ bundle install

Specifying MECAB_PATH

In order to use natto, you need to specify an environment variable called MECAB_PATH.

$ find /usr/ -name "*mecab*" | grep dylib 
$ export MECAB_PATH=/usr//local/Cellar/mecab/0.996/lib/libmecab.dylib

http://yatta47.hateblo.jp/entry/2015/12/13/150525 https://github.com/buruzaemon/natto

Use of mecab

It is OK if it works with the following sample file.

sample.rb

require 'natto'
text = 'Of the thighs and thighs'
nm = Natto::MeCab.new
nm.parse(text) do |n|
  puts "#{n.surface}\t#{n.feature}"
end

http://qiita.com/shizuma/items/d04facaa732f606f00ff http://d.hatena.ne.jp/otn/20090509

3. Original dictionary

It should be made originally, but omitted this time.

This time, instead, we exclude nouns, general pronouns and non-independence.

cond1 = features.include?('noun')
cond2 = features.include?('General')
cond3 = !features.include?('Pronoun')
cond4 = !features.include?('Non-independent')
if cond1 && cond2 && cond3 && cond4
  #Required processing
end

4. Source code from scraping to data shaping

Purpose

Data is exchanged between python and ruby using json. Specifically, prepare a csv that summarizes the URL of the target page as shown below, scrape it from there, and convert it to the data structure required for LDA.

Finally, the following array with words arranged for each document is generated and output as json.

[
  ['human', 'interface', 'computer'],
  ['survey', 'user', 'computer', 'system', 'response', 'time'],
  ['eps', 'user', 'interface', 'system'],
  ['system', 'human', 'system', 'eps'],
  ['user', 'response', 'time'],
  ['trees'],
  ['graph', 'trees'],
  ['graph', 'minors', 'trees'],
  ['graph', 'minors', 'survey']
]

http://tohka383.hatenablog.jp/entry/20111205/1323071336 http://peaceandhilightandpython.hatenablog.com/entry/2013/12/06/082106

Actual source code

gem 'mechanize'
gem 'natto'

#A class that generates an array of URLs from csv
class UrlGetService
  require 'csv'

  def initialize(csv_path)
    @csv_path = csv_path
  end

  def web_urls
    @web_urls ||= -> do
      rows = []
      csv_file.each_with_index do |row, index|
        unless index == 0
          rows << row[0]
        end
      end
      rows
    end.call
  end

  private

    attr_reader :csv_path

    def csv_file
      @csv_file ||= -> do
        csv_text = File.read(csv_path)
        CSV.parse(csv_text)
      end.call
    end
end

#A class that scrapes a given URL
class WebScrapingService
  require 'mechanize'

  def initialize(url)
    @url = url
  end

  def texts
    @texts ||= -> do
      texts = ''
      page_contents.each do |content|
        texts += content.text
      end
      texts
    end.call
  end

  private

    attr_reader :url

    def page_contents
      @page_contents ||= scraping_agent.get(url).search('body p')
    end

    def scraping_agent
      @scraping_agent ||= Mechanize.new
    end
end

#A class that morphologically parses scraping results and creates an array of words
class MorphologicalAnalysisService
  require 'natto'
  `export MECAB_PATH=/usr//local/Cellar/mecab/0.996/lib/libmecab.dylib`

  def initialize(texts)
    @texts = texts
  end

  def words
    words = []
    morphological_analysis_agent.parse(texts) do |word|
      features = word.feature.split(/,/)
      cond1 = features.include?('noun')
      cond2 = features.include?('General')
      cond3 = !features.include?('Pronoun')
      cond4 = !features.include?('Non-independent')
      if cond1 && cond2 && cond3 && cond4
        words << word.surface
      end
    end
    words
  end

  private

    attr_reader :texts

    def morphological_analysis_agent
      @morphological_analysis_agent ||= Natto::MeCab.new
    end
end

#Class that dumps JSON using 3 classes
class DictionaryOutputService
  require 'json'

  def initialize(csv_path)
    @csv_path = csv_path
  end

  def output_json
    open('sample.json', 'w') do |f|
      JSON.dump(words_array, f)
    end
  end

  private

    attr_reader :csv_path

    def words_array
      @words_array ||= -> do
        web_urls.each_with_object([]) do |url, arr|
          texts = WebScrapingService.new(url).texts
          words = MorphologicalAnalysisService.new(texts).words
          white_lists =  words.inject(Hash.new(0)) { |h, a| h[a] += 1; h }.select { |_, c| c > 1 }.map { |w, _| w }
          arr << words.select { |w| white_lists.include?(w) }
        end
      end.call
    end

    def web_urls
      UrlGetService.new(csv_path).web_urls
    end
end

#Execute as follows
csv_path = "YOUR_CSV_PATH/file_name.csv"
DictionaryOutputService.new(csv_path).output_json

5. LDA execution

Version control with pyenv

Instead of using the system python as it is, use the installed and versioned python.

git clone https://github.com/yyuu/pyenv.git ~/.pyenv

~/.bashrc

export PYENV_ROOT=$HOME/.pyenv
export PATH=$PYENV_ROOT/bin:$PATH
eval "$(pyenv init -)"

If it is 3.5 series, you can not fall by installing gensim.

sourve ~/.bashrc
pyenv install 3.5.0
pyenv shell 3.5.0

http://qiita.com/Kodaira_/items/feadfef9add468e3a85b

gensim installation

To do LDA with python, use a module called gensim. setuptools required for gensim installation

sudo easy_install -U setuptools

Install gensim. Also update dependent tools such as numpy.

sudo -H pip install gensim -U

Source code

lda.py

from gensim import models, corpora

if __name__ == '__main__':
    #Originally, this texts reads JSON files etc.
    texts = [['human', 'interface', 'computer'],
             ['survey', 'user', 'computer', 'system', 'response', 'time'],
             ['eps', 'user', 'interface', 'system'],
             ['system', 'human', 'system', 'eps'],
             ['user', 'response', 'time'],
             ['trees'],
             ['graph', 'trees'],
             ['graph', 'minors', 'trees'],
             ['graph', 'minors', 'survey']]

    dictionary = corpora.Dictionary(texts)
    corpus = [dictionary.doc2bow(text) for text in texts]

    lda = models.ldamodel.LdaModel(corpus=corpus, num_topics=20, id2word=dictionary)

    # Topics
    for topic in lda.show_topics(-1):
        print('topic')
        print(topic)

    # Topic of each document
    for topics_per_document in lda[corpus]:
            print('topic of ecah document')
            print(topics_per_document)

https://radimrehurek.com/gensim/tut1.html#corpus-formats https://openbook4.me/projects/193/sections/1154 http://sucrose.hatenablog.com/entry/2013/10/29/001041

Reference: Run LDA on R

#Required Package Army
install.packages("lda")
install.packages("ggplot2")
install.packages("reshape2")

#Free data
data(cora.documents)
data(cora.vocab)

##Number of topics
K <- 10

#Function execution
result <- lda.collapsed.gibbs.sampler(cora.documents,
                                      K, #Number of topics
                                      cora.vocab,
                                      25, #Number of samplings
                                      0.1, #Hyper parameter α
                                      0.1, #Hyper parameter β
                                      compute.log.likelihood=TRUE) 


#Top 5 Frequent Words by Topic
top.words <- top.topic.words(result$topics, 5, by.score=TRUE)