Manifestos Word Clouds

manifestos-word-clouds.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Manifestos Word Clouds",
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/Robin-Lord/c19feef34186c64442e69f6a60ddc516/manifestos-word-clouds.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2uQzUJkJC9wE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Importing libraries to read our pdf\n",
        "\n",
        "import io\n",
        "from pdfminer.converter import TextConverter\n",
        "from pdfminer.pdfinterp import PDFPageInterpreter\n",
        "from pdfminer.pdfinterp import PDFResourceManager\n",
        "from pdfminer.pdfpage import PDFPage\n",
        "\n",
        "\n",
        "# Importing library to show progress bar\n",
        "from tqdm import tqdm_notebook as tqdm\n",
        "\n",
        "# Library to read image and turn it into a mask\n",
        "from PIL import Image\n",
        "import numpy as np\n",
        "\n",
        "# Importing library to construct word clouds and our stopwords list\n",
        "from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MKOwALsyDHLn",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Function to extract text from pdf\n",
        "def extract_text_from_pdf(pdf_path):\n",
        "    resource_manager = PDFResourceManager()\n",
        "    fake_file_handle = io.StringIO()\n",
        "    converter = TextConverter(resource_manager, fake_file_handle)\n",
        "    page_interpreter = PDFPageInterpreter(resource_manager, converter)\n",
        " \n",
        "    with open(pdf_path, 'rb') as fh:\n",
        "        for page in tqdm(PDFPage.get_pages(fh, \n",
        "                                      caching=True,\n",
        "                                      check_extractable=True)):\n",
        "            page_interpreter.process_page(page)\n",
        " \n",
        "        text = fake_file_handle.getvalue()\n",
        "     \n",
        "    print (\"Completed page loop\")\n",
        "    # close open handles\n",
        "    converter.close()\n",
        "    fake_file_handle.close()\n",
        " \n",
        "    if text:\n",
        "        return text"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UYX4fQ6UDCae",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Settings for this word cloud\n",
        "\n",
        "maniesto_file=r\"\"\"Real-Change-Labour-Manifesto-2019.pdf\"\"\"\n",
        "party_name=\"Labour\"\n",
        "party_logo=r\"\"\"labour rose bigger.jpg\"\"\"\n",
        "my_background_colour=\"black\"\n",
        "outfile=party_name+\"_\"+my_background_colour+\".png\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gc1lxrmmDPSg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "resulting_text=extract_text_from_pdf(maniesto_file)\n",
        "\n",
        "# Checking we were able to extract anything\n",
        "print (len(resulting_text))\n",
        "\n",
        "# Joining the text together and printing a sample\n",
        "joined_text=\"\".join(resulting_text)\n",
        "print(joined_text[1800:1900])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_PP8i5p2DQkM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Adding our stopwords (in this case we're also removing the additional words)\n",
        "\n",
        "stopwords_plus=list(STOPWORDS)+[\"will\",\"people\"]\n",
        "print(stopwords_plus)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hhYfSIgMDSXQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "image_to_use=party_logo\n",
        "\n",
        "canvas_and_colours = np.array(Image.open(image_to_use))\n",
        "image_colors = ImageColorGenerator(canvas_and_colours)\n",
        "\n",
        "wordcloud = WordCloud(\n",
        "        background_color=my_background_colour,\n",
        "        mask=canvas_and_colours,\n",
        "        color_func=image_colors,\n",
        "        max_font_size=800,\n",
        "        max_words=3000,\n",
        "    stopwords=stopwords_plus,\n",
        "        margin=0,\n",
        "        random_state=42,\n",
        "        relative_scaling=0.5,\n",
        "    collocations=True).generate(joined_text)\n",
        "\n",
        "wordcloud.to_file(outfile)"
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}