LDA-Wikipedia-Example.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Wikipedia/LDA Demonstration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from nltk.stem.porter import PorterStemmer\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import RegexpTokenizer\n",
    "from gensim import corpora, models\n",
    "import wikipedia\n",
    "import gensim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Scraping Random Wikipedia Articles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "wikipedia_random_articles = wikipedia.random(5)\n",
    "wikipedia_random_articles.append('Car')\n",
    "wikipedia_random_articles.append('Bus')\n",
    "wikipedia_random_articles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "wikipedia_articles = []\n",
    "for wikipedia_article in wikipedia_random_articles:\n",
    "    wikipedia_articles.append([wikipedia_article, wikipedia.page(wikipedia_article).content])\n",
    "\n",
    "    \n",
    "wikipedia_articles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparing the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def clean(article):\n",
    "    title, document = article\n",
    "    tokens = RegexpTokenizer(r'\\w+').tokenize(document.lower())\n",
    "    tokens_clean = [token for token in tokens if token not in stopwords.words('english')]\n",
    "    tokens_stemmed = [PorterStemmer().stem(token) for token in tokens_clean]\n",
    "    return (title, tokens_stemmed)\n",
    "\n",
    "wikipedia_articles_clean = list(map(clean, wikipedia_articles))\n",
    "wikipedia_articles_clean   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Document-Term Matrix and Vectorization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "article_contents = [article[1] for article in wikipedia_articles_clean]\n",
    "dictionary = corpora.Dictionary(article_contents)\n",
    "type(dictionary)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LDA Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "corpus = [dictionary.doc2bow(article) for article in article_contents[:-1]] # All except 'Bus'\n",
    "lda_model = gensim.models.ldamodel.LdaModel(corpus, num_topics=5, id2word = dictionary, passes=100)\n",
    "print(lda_model.print_topics(num_topics=5, num_words=3))\n",
    "\n",
    "# Bus\n",
    "print('\\nBus')\n",
    "print( list( lda_model[ [dictionary.doc2bow(article_contents[-1])] ]) )"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:python35]",
   "language": "python",
   "name": "conda-env-python35-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}