sitc/ml21/preprocessing/09_String_Data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "![](images/EscUpmPolit_p.gif \"UPM\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "# Course Notes for Learning Intelligent Systems"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "Department of Telematic Engineering Systems, Universidad Politécnica de Madrid, © Carlos A. Iglesias"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "## [Introduction to  Preprocessing](00_Intro_Preprocessing.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# String Data\n",
    "It is widespread to clean string columns to follow a predefined format (e.g., emails, URLs, ...).\n",
    "\n",
    "We can do it using regular expressions or specific libraries."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Beautifier\n",
    "A simple [library](https://github.com/labtocat/beautifier) to cleanup and prettify URL patterns, domains, and so on. The library helps to clean Unicode, special characters, and unnecessary redirection patterns from the URLs and gives you a clean date.\n",
    "\n",
    "Install with **'pip install beautifier'**."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Email cleanup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "from beautifier import Email\n",
    "email = Email('me@imsach.in')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'imsach.in'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email.domain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'me'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email.username"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email.is_free_email"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "email2 = Email('This my address')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email2.is_valid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "email3 = Email('pepe@gmail.com')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email3.is_valid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "email3.is_free_email"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## URL cleanup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [],
   "source": [
    "from beautifier import Url\n",
    "url = Url('https://in.linkedin.com/in/sachinphilip?authtoken=887nasdadasd6hasdtg21&secret=98jy766yhhuhnjk')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'https://in.linkedin.com/in/sachinphilip'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.cleanup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'in.linkedin.com'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.domain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['authtoken=887nasdadasd6hasdtg21', 'secret=98jy766yhhuhnjk']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.param"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'authtoken=887nasdadasd6hasdtg21&secret=98jy766yhhuhnjk'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'sachinphilip'"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url.username"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Unicode\n",
    "Problem: Some unicode code has been broken. We see the character in a different character dataset.\n",
    "\n",
    "A **mojibake** is a character displayed in an unintended character encoding. Example:  \"<22>\").\n",
    "\n",
    "We will use the library **ftfy** (fixed text for you) to fix it.\n",
    "\n",
    "First, you should install the library: **conda install ftfy** (or **pip install ftfy**)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "¯\\_(ツ)_/¯\n",
      "Party\n",
      "I'm\n"
     ]
    }
   ],
   "source": [
    "import ftfy\n",
    "foo = '&macr;\\\\_(ã\\x83\\x84)_/&macr;'\n",
    "bar = '\\ufeffParty'\n",
    "baz = '\\001\\033[36;44mI&#x92;m'\n",
    "print(ftfy.fix_text(foo))\n",
    "print(ftfy.fix_text(bar))\n",
    "print(ftfy.fix_text(baz))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "subslide"
    }
   },
   "source": [
    "We can understand which heuristics ftfy is using."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "U+0026  &       [Po] AMPERSAND\n",
      "U+006D  m       [Ll] LATIN SMALL LETTER M\n",
      "U+0061  a       [Ll] LATIN SMALL LETTER A\n",
      "U+0063  c       [Ll] LATIN SMALL LETTER C\n",
      "U+0072  r       [Ll] LATIN SMALL LETTER R\n",
      "U+003B  ;       [Po] SEMICOLON\n",
      "U+005C  \\       [Po] REVERSE SOLIDUS\n",
      "U+005F  _       [Pc] LOW LINE\n",
      "U+0028  (       [Ps] LEFT PARENTHESIS\n",
      "U+00E3  ã       [Ll] LATIN SMALL LETTER A WITH TILDE\n",
      "U+0083  \\x83    [Cc] <unknown>\n",
      "U+0084  \\x84    [Cc] <unknown>\n",
      "U+0029  )       [Pe] RIGHT PARENTHESIS\n",
      "U+005F  _       [Pc] LOW LINE\n",
      "U+002F  /       [Po] SOLIDUS\n",
      "U+0026  &       [Po] AMPERSAND\n",
      "U+006D  m       [Ll] LATIN SMALL LETTER M\n",
      "U+0061  a       [Ll] LATIN SMALL LETTER A\n",
      "U+0063  c       [Ll] LATIN SMALL LETTER C\n",
      "U+0072  r       [Ll] LATIN SMALL LETTER R\n",
      "U+003B  ;       [Po] SEMICOLON\n"
     ]
    }
   ],
   "source": [
    "ftfy.explain_unicode(foo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Dates\n",
    "Sometimes we want to extract date from text. We can use regular expressions or handy packages, such as [**python-dateutil**](https://dateutil.readthedocs.io/en/stable/). An alternative is [arrow](https://arrow.readthedocs.io/en/latest/).\n",
    "\n",
    "Install the library: **pip install python-dateutil**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2019-08-22 10:22:46+00:00\n"
     ]
    }
   ],
   "source": [
    "from dateutil.parser import parse\n",
    "now = parse(\"Thu Aug 22 10:22:46 UTC 2019\")\n",
    "print(now)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2019-08-08 10:20:00\n"
     ]
    }
   ],
   "source": [
    "dt = parse(\"Today is Thursday 8, 2019 at 10:20:00AM\", fuzzy=True)\n",
    "print(dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "# References\n",
    "* [Cleaning and Prepping Data with Python for Data Science — Best Practices and Helpful Packages](https://medium.com/@rrfd/cleaning-and-prepping-data-with-python-for-data-science-best-practices-and-helpful-packages-af1edfbe2a3), DeFilippi, 2019, \n",
    "* [Data Preprocessing for Machine learning in Python, GeeksForGeeks](https://www.geeksforgeeks.org/data-preprocessing-machine-learning-python/), , A. Sharma, 2018.\n",
    "* [Beautifier](https://github.com/labtocat/beautifier) package\n",
    "* [Ftfy](https://ftfy.readthedocs.io/en/latest/) package\n",
    "* [python-dateutil](https://dateutil.readthedocs.io/en/stable/)package"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "skip"
    }
   },
   "source": [
    "## Licence\n",
    "The notebook is freely licensed under under the [Creative Commons Attribution Share-Alike license](https://creativecommons.org/licenses/by/2.0/).  \n",
    "\n",
    "© Carlos A. Iglesias, Universidad Politécnica de Madrid."
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "datacleaner": {
   "position": {
    "top": "50px"
   },
   "python": {
    "varRefreshCmd": "try:\n    print(_datacleaner.dataframe_metadata())\nexcept:\n    print([])"
   },
   "window_display": false
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}