fuzzy matching algorithm.ipynb

{
 "cells": [
  {
   "attachments": {
    "d02cfc97-baa8-4955-919e-2953cad9d501.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ![image.png](attachment:d02cfc97-baa8-4955-919e-2953cad9d501.png)"
   ]
  },
  {
   "attachments": {
    "1f79c007-5587-4999-a03a-e8b746184efe.png": {
     "image/png": "iVBORw0KGgoAAAANSUhEUgAAAKQAAABDCAYAAAAS9K57AAAGHUlEQVR4Ae2auwsdRRSH80fYWdpZ2tlZ2qWzFCy0CYJFbLSw0CaNCAoGRIVAiEXQICqkiAgRVFQQUwRCLBJEkYAEIRBEWPku/G4Ok9ndmbu7N/M4A5d9zZw5j2/PPPaeGLy4BwrywImCdHFV3AODA+kQFOUBB7KocLgyDqQzUJQHHMiiwuHKOJDOQFEecCCLCocr40A6A0V5wIEsKhyujAPpDBTlAQeyqHC4Mg6kM1CUBxzIosLhyjiQzkBRHnAgiwqHK+NAOgNFeeCRAfnX3XvDt9d/Hz7++trw6fc3due37/xTlHNcmeN74KhAAuE7X/w0PPXqueGxF96L/nh2+tw3A3VTC3VPnvls97t2605SM9vG9nXl11t7WZI5d1T7MZkpCl24en3X74tnL0erI/v1C1ejz1Juyq4p+XN2xp6n9J1T52hAvnnxuyiAT792Pgro4y+9P9Dm/r//zdpDZhXgZN2UYtvYzAwYkpV6VPsxmSn6nLn0w65fXshYkWx8ckiRXXPyU21WvUN0mWqzOZB3793fvfky4IlTH+xAU1axyuF0sgB1VP/Zty7OZksFizZLgSTDAof9vfzhlb0+9r7OsZFi9eA8pyAL/VOAAa7ckgMkMZBtc8dcPebqbwokgcLBgouhOCXjUYehRe2QoaDHDLIgLAUyJh+Z0iX2XPesHlsCyejBEJxTcoDM1T1Hj7m6mwL5/Ltf7QPJ4iW3nL38y749mXIMZgtCD0DycjCKpM6X8Xv3QAKgsgpp/9Bi554AGiu9AUmGxLeMHKnZrHsgWazgNFZmSwsykPXkKx9Fs2RvQPKya579zBufRH0S+rxrIO2cK3UIDR1or608HBuW3oDEH/yUKZ97+/PQJQ9ddw2kFiS8vWsVZdzYPlqPQOJXPihoWsSCcap0DSQg4qglc8fQuWxFIDO2LdIrkPiIDw2CkvOxkgNkc9s+mt8csrIec6hdcYd1egYSX5AdBSVZM1ZygJSslGOsryX3Vt/2YcNbhuTulU0ZIociOyy9A4k/mEfiG+aVP978M3RR1rYPW2yxz4Sxew91tPDG6kCyVygg18yQdhsptNmBHHYrbU2VGKHwiS16oWNTHupZH4ZtrZytz1cHEoU1ZDMXWatoP5Ktn7BYZ6au6m2buQDYVX7Yt73OkWnbcc58mxc5BZgxGxmdaC859utW10Bq35DjWkVD0phMZeWxYIV65MBTC5DYeOOPv/cJwX7d6hpIve1kyrHPfSEgU9cWnrGvNQ7kAw/yAmmPks+3lK6B5BurAJnbH3vgxvEzbfng5Ni/hGip/nB8SmHirzZjMiWnpgwpne0eJf7rGkicYv+ylTqMypn2aGGYmpNq4zx171MBis1Jbf+cWx3CZ/baZnLOc4pGlSVzyLA/u0epaVSK/Fzdw36XXG+yqEEhsg7BJgvhhLksFDMCx0gGxykZ+mcRYKYUrUg1pE21qRVIbLJ7lIpFzNYlL1NM3qH3NgMShQik5jLMJ1OHU9qyh6nVOjLm/mr15c+/7YdgzqeKHc7m6soODe9TcpcEdYsMKV21IOweSBxC8AUlDsE5rATHCnM7fQunPlACdkpR1qPd2NDNokj6UD+l1JwhsY+FpXyTMmQTM2xO+aX4L6fOphlSipDdNMcDFn5AwbyGeSHwcB7W4XoKXsnXkQwlx6sPXgDkM6e18gkM9VPKIUDKzqmj7XvLDEk/THewOQXIKZ3DZ9aGNc6PAiSK8paSnTQMh4bZa+aLY9s7c0azGcy8SVnQytU588ZUGOmvBSCxg5d7bFSw0w35KeU4F4/c50cD0ipGxmQFSHYkMypTcm9urmjlTJ2TEWwfTAPIQmvJn+rbnx3ugUcC5OHqesvWPeBAth7hyuxzICsLWOvqOpCtR7gy+xzIygLWuroOZOsRrsw+B7KygLWurgPZeoQrs8+BrCxgravrQLYe4crscyArC1jr6jqQrUe4MvscyMoC1rq6DmTrEa7MPgeysoC1rq4D2XqEK7PPgawsYK2r60C2HuHK7HMgKwtY6+o6kK1HuDL7HMjKAta6ug5k6xGuzD4HsrKAta6uA9l6hCuzz4GsLGCtq/s/fUlYYhSqWSUAAAAASUVORK5CYII="
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:1f79c007-5587-4999-a03a-e8b746184efe.png)\n",
    "\n",
    "- [Introduction to fuzzy matching](#inro)\n",
    "- [Applications of fuzzy matching](#applications)\n",
    "- [Algorithms used for fuzzy matching](#algo)\n",
    "    - [Levenshtein distance algorithm](#lev)\n",
    "    - [Damerau-Levenshtein distance algorithm](#dlev)\n",
    "    - [Bitmap algorithm](#bitap)\n",
    "    - [n-gram algorithm](#ngram)\n",
    "- [Implementation of fuzzy matching on real data](#real)\n",
    "- [Other algorithms](#others)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'intro'></a>"
   ]
  },
  {
   "attachments": {
    "10f32fe3-2664-4602-8916-41a2a2f7d3bf.png": {
     "image/png": "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"
    },
    "cec83173-d5cd-4689-8e25-93ab4e5c28fd.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:cec83173-d5cd-4689-8e25-93ab4e5c28fd.png)\n",
    "\n",
    "- Method to find strings which match a pattern approximately.\n",
    "- Identifies the likelihood/probability that two records are true match based on some parameters.\n",
    "\n",
    "![image.png](attachment:10f32fe3-2664-4602-8916-41a2a2f7d3bf.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'applications'></a>"
   ]
  },
  {
   "attachments": {
    "ed8a9766-a832-4c2f-b3ab-9b80de80c4e4.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:ed8a9766-a832-4c2f-b3ab-9b80de80c4e4.png)\n",
    "\n",
    "- Spell checker\n",
    "- Deduplication of records\n",
    "- Master data management\n",
    "- Plagiarism detection\n",
    "- Bioinformatics and DNA sequencing\n",
    "- Spam filtering\n",
    "- Content search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'algo'></a>"
   ]
  },
  {
   "attachments": {
    "4e03ed85-7d7f-417c-a226-e2a74a4df25f.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:4e03ed85-7d7f-417c-a226-e2a74a4df25f.png)\n",
    "\n",
    "- Edit distance metric - quantifies how dissimilar two strings are by counting the minimum number of operations required to transform one string into the other\n",
    "    - Levenshtein distance\n",
    "    - Damerau–Levenshtein distance\n",
    "- Bitap algorithm - tells whether a given text contains a substring which is \"approximately equal\" (defined in terms of Levenshtein distance) to a given pattern\n",
    "- n-gram - predicts next item in a sequence of text (in form of a Markov model)"
   ]
  },
  {
   "attachments": {
    "6060930b-ba2b-40f1-9ceb-093610d91562.png": {
     "image/png": "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"
    },
    "886c1ed4-1a6f-4c8e-bd8c-049bdf66edb9.png": {
     "image/png": "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"
    },
    "d3d721d8-6b72-4904-95ea-03497287204e.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:d3d721d8-6b72-4904-95ea-03497287204e.png)\n",
    "\n",
    "measures the number of edits needed to transform one word into another\n",
    "\n",
    "![image.png](attachment:886c1ed4-1a6f-4c8e-bd8c-049bdf66edb9.png)\n",
    "\n",
    "![image.png](attachment:6060930b-ba2b-40f1-9ceb-093610d91562.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'lev'></a>"
   ]
  },
  {
   "attachments": {
    "50d382b8-a530-477e-8b9e-156beb647e30.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:50d382b8-a530-477e-8b9e-156beb647e30.png)\n",
    "\n",
    "Levenshtein distance has the following properties:\n",
    "\n",
    "- It is zero if and only if the strings are equal.\n",
    "- It is at least the difference of the sizes of the two strings.\n",
    "- It is at most the length of the longer string.\n",
    "- Triangle inequality: The Levenshtein distance between two strings is no greater than the sum of their Levenshtein distances from a third string.\n",
    "\n",
    "\n",
    "The Levenshtein distance between two strings a and b is given by $lev_{a,b}(len(a), len(b))$.\n",
    "\n",
    "The Levenshtein distance between two strings 'a' and 'b', $lev_{a,b}(len(a), len(b))$,\n",
    "where $lev_{a,b}(i, j)$ is equal to\n",
    "\n",
    "$$\n",
    "\\begin{array}\n",
    "& max(i, j) {\\quad} if {\\quad} min(i, j)=0 \\\\\n",
    "{\\text{otherwise:}} \\\\\n",
    "\\\\\n",
    "min \\big(lev_{a,b}(i-1, j) + 1, \\\\\n",
    "{\\qquad} lev_{a,b}(i, j-1) + 1,  \\\\\n",
    "{\\qquad} lev_{a,b}(i-1, j-1) + 1_{a_i \\neq b_j} \\big) \n",
    "\\end{array}    \n",
    "$$    \n",
    "where $1_{a_i \\neq b_j}$ is the indicator function,\n",
    "$$\\begin{array}\n",
    "& 1_{a_i \\neq b_j} & = 0, {\\text{when }} a_i = b_j \\\\\n",
    "& = 1, {\\text{otherwise}}\n",
    "\\end{array}$$\n",
    "\n",
    "and $lev_{a,b}(i, j)$ is the distance between the first i characters of 'a' and the first j characters of 'b'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def levdist(s, t):\n",
    "    '''function to calculate the\n",
    "    Levenshtein distance between\n",
    "    two strings in a recursive way'''\n",
    "    \n",
    "    if s == '':\n",
    "        return len(t)\n",
    "    if t == '':\n",
    "        return len(s)\n",
    "    if s[-1] == t[-1]:\n",
    "        cost = 0\n",
    "    else:\n",
    "        cost = 1\n",
    "       \n",
    "    dist = min([levdist(s[:-1], t)+1,\n",
    "               levdist(s, t[:-1])+1, \n",
    "               levdist(s[:-1], t[:-1]) + cost])\n",
    "\n",
    "    return dist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "levdist('irks', 'risk')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Levenshtein distance between  micrsft  and  microsoft corporation  is: 14\n",
      "time taken to calculate Levenshtein distance between  micrsft  and  microsoft corporation  is: 28.52 s\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "st = time.time()\n",
    "s1 = 'micrsft'\n",
    "s2 = 'microsoft corporation'\n",
    "print('Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', levdist(s1, s2))\n",
    "et = time.time()\n",
    "print('time taken to calculate Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', round(et-st, 2), 's')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### This is slow! \n",
    "\n",
    "### Iterative computation using matrix improves the computation time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def iterative_levdist(s, t):\n",
    "    '''function to calculate the\n",
    "    Levenshtein distance between\n",
    "    two strings in an iterative way'''\n",
    "\n",
    "    rows = len(s)+1\n",
    "    cols = len(t)+1\n",
    "    dist = [[0 for x in range(cols)] for x in range(rows)]\n",
    "\n",
    "    # source prefixes can be transformed into empty strings \n",
    "    # by deletions:\n",
    "    for i in range(1, rows):\n",
    "        dist[i][0] = i\n",
    "\n",
    "    # target prefixes can be created from an empty source string\n",
    "    # by inserting the characters\n",
    "    for i in range(1, cols):\n",
    "        dist[0][i] = i\n",
    "        \n",
    "    for col in range(1, cols):\n",
    "        for row in range(1, rows):\n",
    "            if s[row-1] == t[col-1]:\n",
    "                cost = 0\n",
    "            else:\n",
    "                cost = 1\n",
    "            dist[row][col] = min(dist[row-1][col] + 1,      # deletion\n",
    "                                 dist[row][col-1] + 1,      # insertion\n",
    "                                 dist[row-1][col-1] + cost) # substitution\n",
    "\n",
    " \n",
    "    return dist[row][col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iterative_levdist('irks', 'risk')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Levenshtein distance between  micrsft  and  microsoft corporation  is: 14\n",
      "time taken to calculate Levenshtein distance between  micrsft  and  microsoft corporation  is: 0.0 s\n"
     ]
    }
   ],
   "source": [
    "st = time.time()\n",
    "s1 = 'micrsft'\n",
    "s2 = 'microsoft corporation'\n",
    "print('Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', iterative_levdist(s1, s2))\n",
    "et = time.time()\n",
    "print('time taken to calculate Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', round(et-st, 2), 's')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'dlev'></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Damerau-Levenshtein Distance works similar to Levenshtein distance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### To install jellyfish package, pip install jellyfish\n",
    "Refer [here](https://pypi.org/project/jellyfish/) for more information on this package"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Levenshtein distance is:  3\n",
      "Damerau-Levenshtein distance is:  2\n"
     ]
    }
   ],
   "source": [
    "import jellyfish\n",
    "\n",
    "print('Levenshtein distance is: ', jellyfish.levenshtein_distance('irks', 'risk'))\n",
    "print('Damerau-Levenshtein distance is: ', jellyfish.damerau_levenshtein_distance('irks', 'risk'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Damerau-Levenshtein distance between  micrsft  and  microsoft corporation  is: 14\n",
      "time taken to calculate Damerau-Levenshtein distance between  micrsft  and  microsoft corporation  is: 0.0 s\n"
     ]
    }
   ],
   "source": [
    "st = time.time()\n",
    "s1 = 'micrsft'\n",
    "s2 = 'microsoft corporation'\n",
    "print('Damerau-Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', jellyfish.damerau_levenshtein_distance(s1, s2))\n",
    "et = time.time()\n",
    "print('time taken to calculate Damerau-Levenshtein distance between ',\n",
    "      f'{s1}', ' and ', f'{s2}', ' is:', round(et-st, 2), 's')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'bitap'></a>"
   ]
  },
  {
   "attachments": {
    "d0c7cfbc-f0e6-4a67-9df0-f9a889917d64.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:d0c7cfbc-f0e6-4a67-9df0-f9a889917d64.png)\n",
    "\n",
    "- also known as the shift-or, shift-and or Baeza-Yates–Gonnet algorithm\n",
    "- states whether a given text contains a substring which is \"approximately equal\" (defined by Levenshtein distance) to a given pattern.\n",
    "\n",
    "\n",
    "**Input:**\n",
    "\n",
    "Text: womenwhocode\n",
    "\n",
    "Pattern: code\n",
    "\n",
    "**Output:** \n",
    "Pattern found at index: 8\n",
    "\n",
    "**Input:**\n",
    "\n",
    "Text: youareawesome \n",
    "\n",
    "Pattern: youareamazing\n",
    "\n",
    "**Output:** No Match"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "\n",
    "def bitap_search(text, pattern):\n",
    "    \n",
    "    '''function to do a bit-ap search\n",
    "    using bit array\n",
    "    (this is python implementation of the\n",
    "    code in GeekforGeeks)'''\n",
    "    \n",
    "    len_pattern = len(pattern)\n",
    "    \n",
    "    #initializing the bit array to complement of 0\n",
    "    bit_array = [~0] * (sys.maxunicode)\n",
    "    \n",
    "    # R is a variable, initiliazing it to complement of 1\n",
    "    R = ~1\n",
    "    \n",
    "    # taking care of the edge cases\n",
    "    # when the pattern is absent\n",
    "    # or when the pattern is longer than the text\n",
    "    if len_pattern == 0:\n",
    "        return -1\n",
    "\n",
    "    if len_pattern > len(text):\n",
    "        print('Pattern too long!')\n",
    "        return -1\n",
    "\n",
    "    \n",
    "    for i in range(len_pattern):\n",
    "        bit_array[ord(pattern[i])] &= ~(1 << i)\n",
    "\n",
    "    for i in range(len(text)):\n",
    "        R |= bit_array[ord(text[i])]\n",
    "        R <<= 1;\n",
    "        if (R & (1 << len_pattern)) == 0:\n",
    "            return i - len_pattern + 1\n",
    "\n",
    "    return -1\n",
    "\n",
    "\n",
    "\n",
    "def findPattern(t, p):\n",
    "    text = list(t);\n",
    "    pattern = list(p);\n",
    "    index = bitap_search(text, pattern);\n",
    "    if index == -1:\n",
    "        print('No Match')\n",
    "    else:\n",
    "        print('Pattern found at index:', index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pattern found at index: 8\n"
     ]
    }
   ],
   "source": [
    "text = 'womenwhocode'\n",
    "pattern = 'code'\n",
    "\n",
    "findPattern(text, pattern)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No Match\n"
     ]
    }
   ],
   "source": [
    "text = 'youareamazing'\n",
    "pattern = 'youareawesome'\n",
    "\n",
    "findPattern(text, pattern)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'ngram'></a>"
   ]
  },
  {
   "attachments": {
    "444fed96-f05c-49e8-80b3-06c74af83568.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:444fed96-f05c-49e8-80b3-06c74af83568.png)\n",
    "\n",
    "- n-gram: set of values generated from a string by pairing sequentially occurring ‘n’ characters/words\n",
    "- goal: compute probability of a sequence of characters/words or sentence \n",
    "- predicts next item in a sequence of text (in form of a Markov model)"
   ]
  },
  {
   "attachments": {
    "921d674b-470e-4c5e-8359-1db40839d2cb.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PROBABILITY \n",
    "\n",
    "Conditional probability with 2 variables/words:\n",
    "\n",
    "$$ \\large{p(w_1 \\cap w_2) = p(w_1) p(w_2|w_1)}$$\n",
    "\n",
    "Conditional probability with 4 variables/words:\n",
    "\n",
    "$$\\large{p(w_1 \\cap w_2 \\cap w_3 \\cap w_4) = p(w_1) p(w_2|w_1) p(w_3|w_1 \\cap w_2) p(w_4|w_1 \\cap w_2 \\cap w_3) }$$\n",
    "\n",
    "Therefore, joint probability can be calculated using chain rule:\n",
    "\n",
    "$$\\Large{ p(w_1 \\cap w_2 , \\cdots, w_n) = \\prod_{i} p (w_i|w_1 w_2 \\cdots w_{n-1}) }$$\n",
    "\n",
    "### &nbsp;&nbsp;&nbsp; TOO MANY POSSIBLE COMBINATIONS!!\n",
    "\n",
    "\n",
    "![image.png](attachment:921d674b-470e-4c5e-8359-1db40839d2cb.png)\n",
    "\n",
    "### Approximate each component of the product by maximum likelihood estimate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "corpus = ['company name is microsoft', 'company the name is Microsoft', \n",
    "          'company name mcrosft', 'the company is Microsft Co',\n",
    "         'company is Microsoft Corporation', 'the company is name microsoft Corp', \n",
    "          'company MCSFT CO name']\n",
    "\n",
    "clean_corpus = ['The company name is Microsoft Corporation']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['co', 'company', 'corp', 'corporation', 'is', 'mcrosft', 'mcsft', 'microsft', 'microsoft', 'name', 'the']\n"
     ]
    }
   ],
   "source": [
    "vect1 = TfidfVectorizer(analyzer='word', ngram_range=(1, 1))\n",
    "x = vect1.fit_transform(corpus)\n",
    "\n",
    "print(vect1.get_feature_names())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['company is microsft', 'company is microsoft', 'company is name', 'company mcsft co', 'company name is', 'company name mcrosft', 'company the name', 'is microsft co', 'is microsoft corporation', 'is name microsoft', 'mcsft co name', 'name is microsoft', 'name microsoft corp', 'the company is', 'the name is']\n"
     ]
    }
   ],
   "source": [
    "vect2 = TfidfVectorizer(analyzer='word', ngram_range=(3, 3))\n",
    "x = vect2.fit_transform(corpus)\n",
    "\n",
    "print(vect2.get_feature_names())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['co name', 'company is', 'company is microsft', 'company is microsft co', 'company is microsoft', 'company is microsoft corporation', 'company is name', 'company is name microsoft', 'company is name microsoft corp', 'company mcsft', 'company mcsft co', 'company mcsft co name', 'company name', 'company name is', 'company name is microsoft', 'company name mcrosft', 'company the', 'company the name', 'company the name is', 'company the name is microsoft', 'is microsft', 'is microsft co', 'is microsoft', 'is microsoft corporation', 'is name', 'is name microsoft', 'is name microsoft corp', 'mcsft co', 'mcsft co name', 'microsft co', 'microsoft corp', 'microsoft corporation', 'name is', 'name is microsoft', 'name mcrosft', 'name microsoft', 'name microsoft corp', 'the company', 'the company is', 'the company is microsft', 'the company is microsft co', 'the company is name', 'the company is name microsoft', 'the name', 'the name is', 'the name is microsoft']\n"
     ]
    }
   ],
   "source": [
    "vect3 = TfidfVectorizer(analyzer='word', ngram_range=(2, 5))\n",
    "x = vect3.fit_transform(corpus)\n",
    "\n",
    "print(vect3.get_feature_names())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let's implement on this example and compare with Levenshtein distance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import itertools\n",
    "\n",
    "# for fuzzy matching\n",
    "from fuzzy_match import algorithims"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('company name is microsoft', 'The company name is Microsoft Corporation'),\n",
       " ('company the name is Microsoft',\n",
       "  'The company name is Microsoft Corporation'),\n",
       " ('company name mcrosft', 'The company name is Microsoft Corporation'),\n",
       " ('the company is Microsft Co', 'The company name is Microsoft Corporation'),\n",
       " ('company is Microsoft Corporation',\n",
       "  'The company name is Microsoft Corporation'),\n",
       " ('the company is name microsoft Corp',\n",
       "  'The company name is Microsoft Corporation'),\n",
       " ('company MCSFT CO name', 'The company name is Microsoft Corporation')]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make a cartesian combination of the words in the corpus and the correct word\n",
    "\n",
    "cart_list = list(itertools.product(corpus, clean_corpus))\n",
    "cart_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_matches(lst):\n",
    "    \n",
    "    '''function to calculate the Levenshtein\n",
    "    distance, trigram match score and the cosine\n",
    "    similarity matches for the above example'''\n",
    "    \n",
    "    x, y, lscore, tscore, cscore = [], [], [], [], []\n",
    "\n",
    "    for i in range(len(lst)):\n",
    "        x.append(lst[i][0])\n",
    "        y.append(lst[i][1])\n",
    "        \n",
    "        lscore.append(round(algorithims.levenshtein(lst[i][0], lst[i][1]), 3))\n",
    "        tscore.append(round(algorithims.trigram(lst[i][0], lst[i][1]), 3))\n",
    "        cscore.append(round(algorithims.cosine(lst[i][0], lst[i][1]), 3))\n",
    "    \n",
    "    df = pd.DataFrame(list(zip(x, y, lscore, tscore, cscore)),\n",
    "                    columns = ['in_data', 'clean_data', 'lev_score', 'tri_score', 'cosine_score'])\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>in_data</th>\n",
       "      <th>clean_data</th>\n",
       "      <th>lev_score</th>\n",
       "      <th>tri_score</th>\n",
       "      <th>cosine_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>company name is microsoft</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.585</td>\n",
       "      <td>0.650</td>\n",
       "      <td>0.612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>company the name is Microsoft</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.512</td>\n",
       "      <td>0.750</td>\n",
       "      <td>0.730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>company name mcrosft</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.463</td>\n",
       "      <td>0.386</td>\n",
       "      <td>0.471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>the company is Microsft Co</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.610</td>\n",
       "      <td>0.512</td>\n",
       "      <td>0.365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>company is Microsoft Corporation</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.780</td>\n",
       "      <td>0.775</td>\n",
       "      <td>0.816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>the company is name microsoft Corp</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.634</td>\n",
       "      <td>0.780</td>\n",
       "      <td>0.500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>company MCSFT CO name</td>\n",
       "      <td>The company name is Microsoft Corporation</td>\n",
       "      <td>0.293</td>\n",
       "      <td>0.333</td>\n",
       "      <td>0.408</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              in_data  \\\n",
       "0           company name is microsoft   \n",
       "1       company the name is Microsoft   \n",
       "2                company name mcrosft   \n",
       "3          the company is Microsft Co   \n",
       "4    company is Microsoft Corporation   \n",
       "5  the company is name microsoft Corp   \n",
       "6               company MCSFT CO name   \n",
       "\n",
       "                                  clean_data  lev_score  tri_score  \\\n",
       "0  The company name is Microsoft Corporation      0.585      0.650   \n",
       "1  The company name is Microsoft Corporation      0.512      0.750   \n",
       "2  The company name is Microsoft Corporation      0.463      0.386   \n",
       "3  The company name is Microsoft Corporation      0.610      0.512   \n",
       "4  The company name is Microsoft Corporation      0.780      0.775   \n",
       "5  The company name is Microsoft Corporation      0.634      0.780   \n",
       "6  The company name is Microsoft Corporation      0.293      0.333   \n",
       "\n",
       "   cosine_score  \n",
       "0         0.612  \n",
       "1         0.730  \n",
       "2         0.471  \n",
       "3         0.365  \n",
       "4         0.816  \n",
       "5         0.500  \n",
       "6         0.408  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_matches(cart_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'real'></a>"
   ]
  },
  {
   "attachments": {
    "41f29434-62b8-4aa7-a289-50803e332989.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:41f29434-62b8-4aa7-a289-50803e332989.png)\n",
    "\n",
    "### Download data [here from Kaggle](https://www.kaggle.com/leandrodoze/room-type).\n",
    "\n",
    "The data contains two columns for room type descriptions. Column 1 is the description from Expedia, and column 2 is the associated room type in Booking.com.\n",
    "\n",
    "<u> **Aim:**</u> is to compare and match these two columns and the result would be 'human like understanding that the matched entries are same'.\n",
    "\n",
    "Packages used: \n",
    "- fuzzywuzzy (Refer [here](https://pypi.org/project/fuzzywuzzy/) for more information).\n",
    "- fuzzy_match (Refer [here](https://pypi.org/project/fuzzy-match/) for more information)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import libraries\n",
    "import pandas as pd\n",
    "\n",
    "# this is for fuzzy matching\n",
    "from fuzzywuzzy import fuzz\n",
    "from fuzzy_match import algorithims"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_df = pd.read_csv('room_type.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Expedia</th>\n",
       "      <th>Booking.com</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Deluxe Room, 1 King Bed</td>\n",
       "      <td>Deluxe King Room</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Standard Room, 1 King Bed, Accessible</td>\n",
       "      <td>Standard King Roll-in Shower Accessible</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Grand Corner King Room, 1 King Bed</td>\n",
       "      <td>Grand Corner King Room</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Suite, 1 King Bed (Parlor)</td>\n",
       "      <td>King Parlor Suite</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>High-Floor Premium Room, 1 King Bed</td>\n",
       "      <td>High-Floor Premium King Room</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 Expedia  \\\n",
       "0                Deluxe Room, 1 King Bed   \n",
       "1  Standard Room, 1 King Bed, Accessible   \n",
       "2     Grand Corner King Room, 1 King Bed   \n",
       "3             Suite, 1 King Bed (Parlor)   \n",
       "4    High-Floor Premium Room, 1 King Bed   \n",
       "\n",
       "                               Booking.com  \n",
       "0                         Deluxe King Room  \n",
       "1  Standard King Roll-in Shower Accessible  \n",
       "2                   Grand Corner King Room  \n",
       "3                        King Parlor Suite  \n",
       "4             High-Floor Premium King Room  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RATIO - Compares the entire string similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_ratio(row):\n",
    "    \n",
    "    '''function to compare the values in each row\n",
    "    for the two columns in the same dataframe and\n",
    "    return the ratio for the entire string similarity'''\n",
    "    \n",
    "    col1 = row['Expedia']\n",
    "    col2 = row['Booking.com']\n",
    "    \n",
    "    return fuzz.ratio(col1, col2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PARTIAL RATIO - Compares partial string similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_partial_ratio(row):\n",
    "    \n",
    "    '''function to compare the values in each row\n",
    "    for the two columns in the same dataframeand\n",
    "    return the ratio for partial string similarity'''\n",
    "    \n",
    "    col1 = row['Expedia']\n",
    "    col2 = row['Booking.com']\n",
    "    \n",
    "    return fuzz.partial_ratio(col1, col2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TOKEN SORT RATIO - Ignores word order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_token_sort_ratio(row):\n",
    "    \n",
    "    '''function to compare the values in each row\n",
    "    for the two columns in the same dataframeand\n",
    "    return the ratio for string similarity by\n",
    "    ignoring word order'''\n",
    "    \n",
    "    col1 = row['Expedia']\n",
    "    col2 = row['Booking.com']\n",
    "    \n",
    "    return fuzz.token_sort_ratio(col1, col2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TOKEN SET RATIO - Ignore duplicate words similarly to token sort ratio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_token_set_ratio(row):\n",
    "    \n",
    "    '''function to compare the values in each row\n",
    "    for the two columns in the same dataframeand\n",
    "    return the ratio for string similarity by\n",
    "    ignoring duplicate words and word order'''\n",
    "    \n",
    "    col1 = row['Expedia']\n",
    "    col2 = row['Booking.com']\n",
    "    \n",
    "    return fuzz.token_set_ratio(col1, col2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### TRIGRAM - Calculates a similarity score and find matches by splitting strings into ngrams with a length of 3. The length of the ngram can be altered if desired."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_trigram_value(row):\n",
    "    \n",
    "    '''function to compare the values in each row\n",
    "    for the two columns in the same dataframeand\n",
    "    return the ratio for string similarity by\n",
    "    ignoring duplicate words and word order'''\n",
    "    \n",
    "    col1 = row['Expedia']\n",
    "    col2 = row['Booking.com']\n",
    "    \n",
    "    return round(algorithims.trigram(col1, col2), 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Expedia</th>\n",
       "      <th>Booking.com</th>\n",
       "      <th>full_ratio</th>\n",
       "      <th>partial_ratio</th>\n",
       "      <th>token_sort_ratio</th>\n",
       "      <th>token_set_ratio</th>\n",
       "      <th>trigram</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Deluxe Room, 1 King Bed</td>\n",
       "      <td>Deluxe King Room</td>\n",
       "      <td>62</td>\n",
       "      <td>69</td>\n",
       "      <td>84</td>\n",
       "      <td>100</td>\n",
       "      <td>0.739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Standard Room, 1 King Bed, Accessible</td>\n",
       "      <td>Standard King Roll-in Shower Accessible</td>\n",
       "      <td>68</td>\n",
       "      <td>65</td>\n",
       "      <td>78</td>\n",
       "      <td>81</td>\n",
       "      <td>0.562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Grand Corner King Room, 1 King Bed</td>\n",
       "      <td>Grand Corner King Room</td>\n",
       "      <td>79</td>\n",
       "      <td>100</td>\n",
       "      <td>80</td>\n",
       "      <td>100</td>\n",
       "      <td>0.793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Suite, 1 King Bed (Parlor)</td>\n",
       "      <td>King Parlor Suite</td>\n",
       "      <td>51</td>\n",
       "      <td>65</td>\n",
       "      <td>85</td>\n",
       "      <td>100</td>\n",
       "      <td>0.750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>High-Floor Premium Room, 1 King Bed</td>\n",
       "      <td>High-Floor Premium King Room</td>\n",
       "      <td>76</td>\n",
       "      <td>82</td>\n",
       "      <td>90</td>\n",
       "      <td>100</td>\n",
       "      <td>0.829</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 Expedia  \\\n",
       "0                Deluxe Room, 1 King Bed   \n",
       "1  Standard Room, 1 King Bed, Accessible   \n",
       "2     Grand Corner King Room, 1 King Bed   \n",
       "3             Suite, 1 King Bed (Parlor)   \n",
       "4    High-Floor Premium Room, 1 King Bed   \n",
       "\n",
       "                               Booking.com  full_ratio  partial_ratio  \\\n",
       "0                         Deluxe King Room          62             69   \n",
       "1  Standard King Roll-in Shower Accessible          68             65   \n",
       "2                   Grand Corner King Room          79            100   \n",
       "3                        King Parlor Suite          51             65   \n",
       "4             High-Floor Premium King Room          76             82   \n",
       "\n",
       "   token_sort_ratio  token_set_ratio  trigram  \n",
       "0                84              100    0.739  \n",
       "1                78               81    0.562  \n",
       "2                80              100    0.793  \n",
       "3                85              100    0.750  \n",
       "4                90              100    0.829  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_df['full_ratio'] = data_df.apply(get_ratio, axis=1)\n",
    "data_df['partial_ratio'] = data_df.apply(get_partial_ratio, axis=1)\n",
    "data_df['token_sort_ratio'] = data_df.apply(get_token_sort_ratio, axis=1)\n",
    "data_df['token_set_ratio'] = data_df.apply(get_token_set_ratio, axis=1)\n",
    "data_df['trigram'] = data_df.apply(get_trigram_value, axis=1)\n",
    "data_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  It looks like TOKEN SET RATIO from fuzzywuzzy package is the best method to get the most similar matches in this example."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id = 'others'></a>"
   ]
  },
  {
   "attachments": {
    "fda7874a-b7b2-484e-b187-397b35c93cd6.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:fda7874a-b7b2-484e-b187-397b35c93cd6.png)\n",
    "\n",
    "- Edit distance \n",
    "    - Longest common subsequence \n",
    "    - Hamming distance\n",
    "    - Jaro distance\n",
    "- Needleman–Wunsch algorithm\n",
    "- Smith–Waterman algorithm\n",
    "- BK Tree metric\n",
    "- Soundex or Metaphone – phonetic algorithms \n",
    "\n",
    "[Record Linkange Toolkit](https://recordlinkage.readthedocs.io/en/latest/about.html) library to link records in or between data sources and provides tools for deduplication and record linkage."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# REFERENCES\n",
    "1. [Levenshtein, Vladimir I. \"Binary codes capable of correcting deletions, insertions, and reversals.\" In Soviet physics doklady, vol. 10, no. 8, pp. 707-710. 1966.](https://nymity.ch/sybilhunting/pdf/Levenshtein1966a.pdf)\n",
    "2. [Damerau, Fred J. \"A technique for computer detection and correction of spelling errors.\" Communications of the ACM 7, no. 3 (1964): 171-176.](https://dl.acm.org/doi/abs/10.1145/363958.363994)\n",
    "3. [Cayrol, M., Farreny, H. and Prade, H. (1982), 'Fuzzy Pattern Matching', Kybernetes, Vol. 11 No. 2, pp. 103-116.](https://doi.org/10.1108/eb005612)\n",
    "4. [Ukkonen, Esko. \"Algorithms for approximate string matching.\" Information and control 64, no. 1-3 (1985): 100-118.](https://reader.elsevier.com/reader/sd/pii/S0019995885800462?token=6B1FB04FFCA70F0F9A2AAB96C6311FAA5E93FFB7F1C1B4A75A61D72728A2B6A655B4944421AD983395F0B11AAB73F520&originRegion=us-east-1&originCreation=20210331153641)\n",
    "5. [Geek for Geeks - applications of fuzzy string matching](https://www.geeksforgeeks.org/applications-of-string-matching-algorithms/)\n",
    "6. [Geek for Geeks - Bitap Algorithm](https://www.geeksforgeeks.org/java-program-to-implement-bitap-algorithm-for-string-matching/)\n",
    "7. [Stanford slides on n-gram](https://web.stanford.edu/~jurafsky/slp3/slides/LM_4.pdf)\n",
    "8. [Data camp tutorial - fuzzy string matching](https://www.datacamp.com/community/tutorials/fuzzy-string-python)\n",
    "9. [Levenshtein distance theory](https://www.python-course.eu/levenshtein_distance.php)\n",
    "10. [Article on record linking and fuzzy matching](https://pbpython.com/record-linking.html)\n",
    "11. [Medium post on Levenshtein distance](https://blog.paperspace.com/measuring-text-similarity-using-levenshtein-distance/)\n",
    "12. [stackoverflow for n-gram similarity](https://stackoverflow.com/questions/53827339/string-matching-using-tf-idf-ngrams-and-cosine-similarity-in-python)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}