Introduction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Topics - Easy Topic Modeling in Python"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The text mining technique Topic Modeling has become a popular statistical method for clustering documents. This notebook introduces an user-friendly workflow, basically containing data preprocessing, an implementation of LDA (Latent Dirichlet Allocation) topic modeling which learns the relationships between words, topics, and documents, as well as a visualization to explore the trained LDA model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Preperations\n",
    "\n",
    "The following tutorial will explain how to perform LDA topic modeling with a programming library in Python. If you have not done so yet, please follow the instructions for installing jupyter and all necessary python libraries mentioned in readme.txt/installation_instructions.\n",
    "\n",
    "Before you start, make sure to have Git and the Topics repository ready to use on your computer:\n",
    "1. Download and install [Git](https://git-scm.com/book/en/v2/Getting-Started-Installing-Git)\n",
    "2. Open the [command-line interface](https://en.wikipedia.org/wiki/Command-line_interface), type `git clone https://github.com/DARIAH-DE/Topics.git` and press Enter\n",
    "3. Access the new folder *topics* in your package explorer\n",
    "4. To install the required packages, simply run `setup.py`\n",
    "5. Install [Jupyter](http://jupyter.readthedocs.io/en/latest/install.html) and run it by typing `jupyter notebook` in the command-line\n",
    "5. Access the folder **Topics** through Jupyter in your browser, open **IntroductionTopics.ipynb** and follow the instructions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## 1. Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.1. Loading modules"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Loading modules from DARIAH-Topics library\n",
    "First, we have to get access to the functionalities of the library by importing them. For using its functions we use the prefix of the toolbox's submodules (pre, visual and mallet)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "ename": "SyntaxError",
     "evalue": "invalid syntax (mallet.py, line 81)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;36m  File \u001b[0;32m\"dariah_topics/mallet.py\"\u001b[0;36m, line \u001b[0;32m81\u001b[0m\n\u001b[0;31m    log.debug(print(param))\u001b[0m\n\u001b[0m                  ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/lib/python2.7/json/encoder.py:207: DeprecationWarning: Interpreting naive datetime as local 2017-03-21 16:50:50.477599. Please add timezone info to timestamps.\n",
      "  chunks = self.iterencode(o, _one_shot=True)\n"
     ]
    }
   ],
   "source": [
    "from dariah_topics import preprocessing as pre\n",
    "from dariah_topics import visualization as visual\n",
    "from dariah_topics import mallet as mal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Load required functions from Gensim\n",
    "Furthermore, we will need some additional functions from external libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from gensim.models import LdaModel\n",
    "from gensim.corpora import MmCorpus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Activating inline output in Jupyter notebook\n",
    "The following line will just tell the notebook to show graphics in the output frames."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.4/dist-packages/traitlets/traitlets.py:809: DeprecationWarning: A parent of InlineBackend._config_changed has adopted the new (traitlets 4.1) @observe(change) API\n",
      "  clsname, change_or_name), DeprecationWarning)\n",
      "/usr/lib/python3.4/importlib/_bootstrap.py:321: DeprecationWarning: `IPython.lib.inputhook` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  return f(*args, **kwds)\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:327: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('osx')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:336: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('wx')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:398: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('qt', 'qt4')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:449: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('qt5')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:458: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('gtk')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:489: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('tk')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:522: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('glut')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:596: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('pyglet')\n",
      "/usr/local/lib/python3.4/dist-packages/IPython/lib/inputhook.py:624: DeprecationWarning: `register` is deprecated since IPython 5.0 and will be removed in future versions.\n",
      "  @inputhook_manager.register('gtk3')\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.2. Reading a corpus of documents"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Defining the path to the corpus folder\n",
    "\n",
    "In the present example code, we are using a folder of 'txt' documents provided with the package. For using your own corpus, change the path accordingly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "path = \"corpus_txt\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### List all documents in the folder\n",
    "We begin by creating a list of all the documents in the folder specified above. That list will tell function `pre.read_from_txt()` (see below) which text documents to read."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doclist = pre.create_document_list(path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The current list of documents looks like this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['corpus_txt/Kipling_TheEndofthePassage.txt',\n",
       " 'corpus_txt/Howard_TheDevilinIron.txt',\n",
       " 'corpus_txt/Poe_TheMasqueoftheRedDeath.txt',\n",
       " 'corpus_txt/Doyle_AScandalinBohemia.txt',\n",
       " 'corpus_txt/Doyle_AStudyinScarlet.txt',\n",
       " 'corpus_txt/Kipling_ThyServantaDog.txt',\n",
       " 'corpus_txt/Poe_TheCaskofAmontillado.txt',\n",
       " 'corpus_txt/Howard_ShadowsintheMoonlight.txt',\n",
       " 'corpus_txt/Howard_GodsoftheNorth.txt',\n",
       " 'corpus_txt/Howard_SchadowsinZamboula.txt',\n",
       " 'corpus_txt/Doyle_TheSignoftheFour.txt',\n",
       " 'corpus_txt/Lovecraft_AttheMountainofMadness.txt',\n",
       " 'corpus_txt/Poe_ThePurloinedLetter.txt',\n",
       " 'corpus_txt/Lovecraft_TheShunnedHouse.txt',\n",
       " 'corpus_txt/Poe_EurekaAProsePoem.txt',\n",
       " 'corpus_txt/Kipling_TheJungleBook.txt',\n",
       " 'corpus_txt/Doyle_TheHoundoftheBaskervilles.txt']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doclist"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "**Alternatively**, if we want to use other documents, or just a selction of those in the specified folder, we can define our own `doclist` by creating a list of strings containing paths to text files. For example, to use only the texts by Edgar A. Poe from the current folder, we would define the list as\n",
    "\n",
    "`\n",
    "    doclist = [ 'corpus_txt/Poe_TheMasqueoftheRedDeath.txt', \n",
    "            'corpus_txt/Poe_TheCaskofAmontillado.txt', \n",
    "            'corpus_txt/Poe_ThePurloinedLetter.txt',\n",
    "            'corpus_txt/Poe_ThePurloinedLetter.txt',\n",
    "            'corpus_txt/Poe_EurekaAProsePoem.txt']\n",
    "`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Generate document labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doc_labels = list(pre.get_labels(doclist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Read listed documents from folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "corpus = pre.read_from_txt(doclist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "At this point, the corpus is generator object."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.3. Tokenize corpus\n",
    "Your text files will be tokenized. Tokenization is the task of cutting a stream of characters into linguistic units, simply words or, more precisely, tokens. The tokenize function the library provides is a simple unicode tokenizer. Depending on the corpus it might be useful to use an external tokenizer function, or even develop your own, since its efficiency varies with language, epoch and text type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doc_tokens = [list(pre.tokenize(doc)) for doc in list(corpus)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "At this point, each text is represented by a list of separate token strings. If we want to look e.g. into the first text (which has the index `0` as Python starts counting at 0) and show its first 10 words/tokens (that have the indeces `0:9` accordingly) by typing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['title', 'at', 'the', 'end', 'of', 'the', 'passage', 'author', 'rudyard']"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc_tokens[0][0:9]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.4. Create a sparse bag-of-words model\n",
    "\n",
    "The LDA topic model is based on a bag-of-words model of the corpus. To improve performance in large corpora, actual words and document titels are replaced by indices in the actual bag-of-words model. It is therefore necessary to create dictionaries for mapping these indices in advance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Create dictionaries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "id_types, doc_ids = pre.create_dictionaries(doc_labels, doc_tokens)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Create market matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sparse_bow = pre.create_mm(doc_labels, doc_tokens, id_types, doc_ids)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.5. Feature selection and/or removal\n",
    "\n",
    "In topic modeling, it is often usefull (if not vital) to remove some types before modeling. In this example, the 100 most frequent words and the *hapax legomena* in the corpus are listed and removed. Alternatively, the 'feature_list' containing all features to be removed from the corpus can be replaced by, or combined with an external stop word list or any other list of strings containing features we want to remove."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### List the 100 most frequent words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "mfw100 = pre.find_stopwords(sparse_bow, id_types, 100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### List hapax legomena"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "hapax_list = pre.find_hapax(sparse_bow, id_types)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Combine lists and remove content from bag-of-words model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "feature_list = set(mfw100).union(hapax_list)\n",
    "clean_term_frequency = pre.remove_features(sparse_bow, id_types, feature_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 1.6. Save bag-of-words model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "pre.save_bow_mm(sparse_bow, \"gb_plain\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## 2. Model creation\n",
    "\n",
    "The actual topic modeling is done with external state-of-the-art LDA implementations. In this example, we are relying on the open-source toolkit **Gensim** which was used and cited in over 400 commercial and academic applications since 2008. As an alternative, the Java program **Mallet** can be used for LDA. Instruction on how to use Mallet can be found in the [Mallet notebook](Mallet.ipynb)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 2.1. Load corpus object"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "mm = MmCorpus(\"gb_plain.mm\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 2.2. Rearrange dictionaries "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doc2id = {value : key for key, value in doc_ids.items()}\n",
    "type2id = {value : key for key, value in id_types.items()}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 2.3. Generate LDA model\n",
    "\n",
    "We can define the number of topics we want to calculate as an argument in the function. Furthermore, the number of passes can be defined. A higher number of passes will probably yield a better model, but also increase processing time.\n",
    "\n",
    "**Warning: this step can take quite a while!** Meaning something between some seconds and some hours depending on corpus size and the number of passes. Our example short stories corpus should be done within a minute or two at 'passes = 10'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "model = LdaModel(corpus=mm, id2word=type2id, num_topics=10, passes = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 2.4. Create document-topic matrix\n",
    "\n",
    "The generated model object can now be translated into a human-readable document-topic matrix (that is a actually a pandas data frame) that constitutes our principle exchange format for topic modeling results. For generating the matrix from a Gensim model, we can use the following function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "doc_topic = visual.create_doc_topic(mm, model, doc_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## 3. Model visualization and evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Now we can see the topics in the model with the following function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Key 1</th>\n",
       "      <th>Key 2</th>\n",
       "      <th>Key 3</th>\n",
       "      <th>Key 4</th>\n",
       "      <th>Key 5</th>\n",
       "      <th>Key 6</th>\n",
       "      <th>Key 7</th>\n",
       "      <th>Key 8</th>\n",
       "      <th>Key 9</th>\n",
       "      <th>Key 10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Topic 1</th>\n",
       "      <td>wadding</td>\n",
       "      <td>streaks</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>ledges</td>\n",
       "      <td>recompense</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>camel</td>\n",
       "      <td>close-knit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 2</th>\n",
       "      <td>wadding</td>\n",
       "      <td>streaks</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>ledges</td>\n",
       "      <td>recompense</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>camel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 3</th>\n",
       "      <td>wadding</td>\n",
       "      <td>streaks</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>recompense</td>\n",
       "      <td>ledges</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>camel</td>\n",
       "      <td>jewels</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 4</th>\n",
       "      <td>wadding</td>\n",
       "      <td>streaks</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>seasonal</td>\n",
       "      <td>agglomerative</td>\n",
       "      <td>ledges</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>jewels</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 5</th>\n",
       "      <td>wadding</td>\n",
       "      <td>leathers</td>\n",
       "      <td>streaks</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>ledges</td>\n",
       "      <td>recompense</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>neolithic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 6</th>\n",
       "      <td>wadding</td>\n",
       "      <td>leathers</td>\n",
       "      <td>nurse</td>\n",
       "      <td>streaks</td>\n",
       "      <td>u-boat</td>\n",
       "      <td>camel</td>\n",
       "      <td>launched</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>crocodile</td>\n",
       "      <td>recently</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 7</th>\n",
       "      <td>wadding</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>streaks</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>impassable</td>\n",
       "      <td>camel</td>\n",
       "      <td>ledges</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 8</th>\n",
       "      <td>wadding</td>\n",
       "      <td>leathers</td>\n",
       "      <td>streaks</td>\n",
       "      <td>camel</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>neolithic</td>\n",
       "      <td>nurse</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 9</th>\n",
       "      <td>wadding</td>\n",
       "      <td>leathers</td>\n",
       "      <td>streaks</td>\n",
       "      <td>isn't</td>\n",
       "      <td>ledges</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>recompense</td>\n",
       "      <td>camel</td>\n",
       "      <td>close-knit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Topic 10</th>\n",
       "      <td>wadding</td>\n",
       "      <td>streaks</td>\n",
       "      <td>leathers</td>\n",
       "      <td>isn't</td>\n",
       "      <td>fox-hounds</td>\n",
       "      <td>inviolate</td>\n",
       "      <td>recompense</td>\n",
       "      <td>close-knit</td>\n",
       "      <td>ledges</td>\n",
       "      <td>jewels</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Key 1     Key 2     Key 3    Key 4       Key 5       Key 6  \\\n",
       "Topic 1   wadding   streaks  leathers    isn't  fox-hounds      ledges   \n",
       "Topic 2   wadding   streaks  leathers    isn't  fox-hounds      ledges   \n",
       "Topic 3   wadding   streaks  leathers    isn't  recompense      ledges   \n",
       "Topic 4   wadding   streaks  leathers    isn't  fox-hounds    seasonal   \n",
       "Topic 5   wadding  leathers   streaks    isn't  fox-hounds  close-knit   \n",
       "Topic 6   wadding  leathers     nurse  streaks      u-boat       camel   \n",
       "Topic 7   wadding  leathers     isn't  streaks   inviolate  fox-hounds   \n",
       "Topic 8   wadding  leathers   streaks    camel   inviolate       isn't   \n",
       "Topic 9   wadding  leathers   streaks    isn't      ledges  fox-hounds   \n",
       "Topic 10  wadding   streaks  leathers    isn't  fox-hounds   inviolate   \n",
       "\n",
       "                  Key 7       Key 8       Key 9      Key 10  \n",
       "Topic 1      recompense   inviolate       camel  close-knit  \n",
       "Topic 2      recompense   inviolate  close-knit       camel  \n",
       "Topic 3      fox-hounds   inviolate       camel      jewels  \n",
       "Topic 4   agglomerative      ledges  close-knit      jewels  \n",
       "Topic 5          ledges  recompense   inviolate   neolithic  \n",
       "Topic 6        launched  fox-hounds   crocodile    recently  \n",
       "Topic 7      close-knit  impassable       camel      ledges  \n",
       "Topic 8      fox-hounds  close-knit   neolithic       nurse  \n",
       "Topic 9       inviolate  recompense       camel  close-knit  \n",
       "Topic 10     recompense  close-knit      ledges      jewels  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "visual.topicwords_in_df(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "### 3.1. Distribution of topics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Distribution of topics over all documents\n",
    "\n",
    "The distribution of topics over all documents can now be visualized in a heat map:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f854635a5f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "heatmap = visual.doc_topic_heatmap(doc_topic)\n",
    "heatmap.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "#### Distribution of topics in a single documents\n",
    "\n",
    "To take closer look on the topics in a single text, we can use the follwing function that shows all the topics in a text and their respective proportions. To select the document, we have to give its index to the function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAakAAAEaCAYAAACrcqiAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ng2VISZIGy5CSJA2WISVJGixDSpI0WP8fOHH1jUJ1D/AAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8545e04b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "visual.plot_doc_topics(doc_topic, 3)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}