Gender_Classification

Gender_classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, we will determine,for a given author or authors, whether there are differences in portrail of characters based on gender. \n",
    "\n",
    "Our final task is to define a function that guesses the gender of a character based on his or her name.  \n",
    "\n",
    "The file `names.csv` contains several thousand male and female names with information about how frequent each name is. This information is extracted online. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Novel Used: Alice in the wonderland"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Function to create gender map from names.csv file\n",
    "def create_gender_map(dict_reader):\n",
    "    names_info = defaultdict(lambda: {\"gender\":\"\", \"freq\": 0.0})\n",
    "    for row in input_file:\n",
    "        name = row[\"name\"].lower()\n",
    "        if names_info[name][\"freq\"] < float(row[\"freq\"]): # is this gender more frequent?\n",
    "            names_info[name][\"gender\"] = row[\"gender\"] \n",
    "            names_info[name][\"freq\"] = float(row[\"freq\"])\n",
    "    gender_map = defaultdict(lambda: \"unknown\")\n",
    "    for name in names_info:\n",
    "        gender_map[name] = names_info[name][\"gender\"]\n",
    "    return gender_map\n",
    "\n",
    "os.chdir('C://Users//cmm//Desktop')\n",
    "\n",
    "input_file = csv.DictReader(open('names.csv')) ## Importing our names.csv file\n",
    "gender_map = create_gender_map(input_file)    ## Import the gender map\n",
    "#### Male homonyms\n",
    "male_title=['mr.','sir','monsieur','captain','chief','master','lord','baron','mister','mr','prince','king']\n",
    "#### Female homonyms\n",
    "female_title=['mrs.','ms.','miss','lady','madameoiselle','baroness','mistress','mrs','ms','queen','princess','madam','madame']\n",
    "\n",
    "\n",
    "def gender_guess(name,gender_map):                   #Identifying entries in the names.csv database#\n",
    "    if (len(name.split()))==1:\n",
    "        if name.lower() in gender_map.keys():\n",
    "            return gender_map[name]\n",
    "        else:\n",
    "            return('unknown')\n",
    "    \n",
    "    if(len(name.split()))>1:                        \n",
    "        name_array=name.lower().split()\n",
    "        if name_array[0] in gender_map.keys():\n",
    "            return gender_map[name_array[0]]\n",
    "        \n",
    "        \n",
    "        for title in name_array:                     #Recognising titles of entries# \n",
    "            if title in male_title:\n",
    "                return 'male'\n",
    "            elif title in female_title:\n",
    "                return 'female'\n",
    "            else: \n",
    "                return('unknown')\n",
    "                break\n",
    "        \n",
    "def named_entity_counts(document,named_entity_label):    \n",
    "    \n",
    "## Function that outputs a Counter object of human entities found\n",
    "\n",
    "    occurrences = [ent.string.strip() for ent in document.ents \n",
    "                   if ent.label_ == named_entity_label and ent.string.strip()]\n",
    "    return Counter(occurrences)\n",
    "\n",
    "alice = gutenberg.raw(fileids='carroll-alice.txt')\n",
    "parsed_alice=nlp(alice)\n",
    "text = parsed_alice    ### Parsing Alice in the wonderland by Lewis Carroll\n",
    "entity_type = 'PERSON' ## Type of entry\n",
    "number_of_entities = 10  ### Control over obtaining number of defined type entities\n",
    "Entities=pd.DataFrame(named_entity_counts(text,entity_type).most_common(number_of_entities),columns=[\"Entity\",\"Count\"])\n",
    "entity=[]\n",
    "for char in Entities['Entity']:\n",
    "    entity.append(gender_guess(char.lower(),gender_map))\n",
    "Entities['Pred_Gender']=entity\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Entity</th>\n",
       "      <th>Count</th>\n",
       "      <th>Pred_Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alice</td>\n",
       "      <td>388</td>\n",
       "      <td>female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Gryphon</td>\n",
       "      <td>32</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Queen</td>\n",
       "      <td>27</td>\n",
       "      <td>female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Mouse</td>\n",
       "      <td>25</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Rabbit</td>\n",
       "      <td>20</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Bill</td>\n",
       "      <td>12</td>\n",
       "      <td>male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Dormouse</td>\n",
       "      <td>8</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Dinah</td>\n",
       "      <td>8</td>\n",
       "      <td>female</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>William</td>\n",
       "      <td>5</td>\n",
       "      <td>male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Beau</td>\n",
       "      <td>4</td>\n",
       "      <td>male</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Entity  Count Pred_Gender\n",
       "0     Alice    388      female\n",
       "1   Gryphon     32     unknown\n",
       "2     Queen     27      female\n",
       "3     Mouse     25     unknown\n",
       "4    Rabbit     20     unknown\n",
       "5      Bill     12        male\n",
       "6  Dormouse      8     unknown\n",
       "7     Dinah      8      female\n",
       "8   William      5        male\n",
       "9      Beau      4        male"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Entities"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluating the Gender classifier for different entities"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Male type Entities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "male\n",
      "male\n",
      "male\n",
      "male\n",
      "male\n",
      "male\n",
      "male\n"
     ]
    }
   ],
   "source": [
    "names=['harry','abdul','homer','gary','robert','wayne','lionel']\n",
    "for name in names:\n",
    "    print(gender_guess(name,gender_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Female type entities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "female\n",
      "female\n",
      "female\n",
      "female\n",
      "female\n",
      "female\n",
      "female\n"
     ]
    }
   ],
   "source": [
    "names=['martha','holly','nicole','catherine','ruth','april','christina']\n",
    "for name in names:\n",
    "    print(gender_guess(name,gender_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### For first and last names given"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "female\n",
      "female\n",
      "male\n",
      "male\n",
      "male\n",
      "male\n"
     ]
    }
   ],
   "source": [
    "names=['Liz Lemon','Leslie Knope','jesus navas','Robert Lewandowski','Anthony Martial','Wesley Sneijder']\n",
    "for name in names:\n",
    "    print(gender_guess(name,gender_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### For names with titles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "male\n",
      "female\n",
      "male\n"
     ]
    }
   ],
   "source": [
    "print(gender_guess('Sir Alex Ferguson',gender_map))\n",
    "print(gender_guess('Lady McElroy',gender_map))\n",
    "print(gender_guess('Friedrich Nietzsche',gender_map))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### The program returns 'unknown' if the gender of the entity can't be determined by the function created. This error is a result of the name not being in the names.csv folder or doesn't have a gender bisecting title attached to it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unknown\n",
      "unknown\n",
      "unknown\n"
     ]
    }
   ],
   "source": [
    "print(gender_guess('Liam Neeson',gender_map))\n",
    "print(gender_guess('Mahatma Gandhi',gender_map))\n",
    "print(gender_guess('Pricella McCartney',gender_map))"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}