ML-K_Means-Clustering-Congress-Voting.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.cluster import KMeans\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=DeprecationWarning)\n",
    "warnings.filterwarnings(\"ignore\", category=FutureWarning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('114_congress.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "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>name</th>\n",
       "      <th>party</th>\n",
       "      <th>state</th>\n",
       "      <th>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alexander</td>\n",
       "      <td>R</td>\n",
       "      <td>TN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ayotte</td>\n",
       "      <td>R</td>\n",
       "      <td>NH</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Baldwin</td>\n",
       "      <td>D</td>\n",
       "      <td>WI</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Barrasso</td>\n",
       "      <td>R</td>\n",
       "      <td>WY</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Bennet</td>\n",
       "      <td>D</td>\n",
       "      <td>CO</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        name party state  00001  00004  00005  00006  00007  00008  00009  \\\n",
       "0  Alexander     R    TN    0.0    1.0    1.0    1.0    1.0    0.0    0.0   \n",
       "1     Ayotte     R    NH    0.0    1.0    1.0    1.0    1.0    0.0    0.0   \n",
       "2    Baldwin     D    WI    1.0    0.0    0.0    1.0    0.0    1.0    0.0   \n",
       "3   Barrasso     R    WY    0.0    1.0    1.0    1.0    1.0    0.0    1.0   \n",
       "4     Bennet     D    CO    0.0    0.0    0.0    1.0    0.0    1.0    0.0   \n",
       "\n",
       "   00010  00020  00026  00032  00038  00039  00044  00047  \n",
       "0    1.0    1.0    1.0    0.0    0.0    0.0    0.0    0.0  \n",
       "1    1.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0  \n",
       "2    1.0    0.0    0.0    1.0    1.0    0.0    1.0    1.0  \n",
       "3    1.0    1.0    1.0    0.0    0.0    1.0    0.0    0.0  \n",
       "4    1.0    0.0    0.0    0.0    1.0    0.0    1.0    0.0  "
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 18)"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100 entries, 0 to 99\n",
      "Data columns (total 18 columns):\n",
      " #   Column  Non-Null Count  Dtype  \n",
      "---  ------  --------------  -----  \n",
      " 0   name    100 non-null    object \n",
      " 1   party   100 non-null    object \n",
      " 2   state   100 non-null    object \n",
      " 3   00001   100 non-null    float64\n",
      " 4   00004   100 non-null    float64\n",
      " 5   00005   100 non-null    float64\n",
      " 6   00006   100 non-null    float64\n",
      " 7   00007   100 non-null    float64\n",
      " 8   00008   100 non-null    float64\n",
      " 9   00009   100 non-null    float64\n",
      " 10  00010   100 non-null    float64\n",
      " 11  00020   100 non-null    float64\n",
      " 12  00026   100 non-null    float64\n",
      " 13  00032   100 non-null    float64\n",
      " 14  00038   100 non-null    float64\n",
      " 15  00039   100 non-null    float64\n",
      " 16  00044   100 non-null    float64\n",
      " 17  00047   100 non-null    float64\n",
      "dtypes: float64(15), object(3)\n",
      "memory usage: 14.2+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.00000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>100.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.325000</td>\n",
       "      <td>0.575000</td>\n",
       "      <td>0.535000</td>\n",
       "      <td>0.94500</td>\n",
       "      <td>0.54500</td>\n",
       "      <td>0.415000</td>\n",
       "      <td>0.54500</td>\n",
       "      <td>0.985000</td>\n",
       "      <td>0.525000</td>\n",
       "      <td>0.545000</td>\n",
       "      <td>0.410000</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.510000</td>\n",
       "      <td>0.460000</td>\n",
       "      <td>0.370000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.462618</td>\n",
       "      <td>0.494286</td>\n",
       "      <td>0.498761</td>\n",
       "      <td>0.22355</td>\n",
       "      <td>0.49795</td>\n",
       "      <td>0.492648</td>\n",
       "      <td>0.49795</td>\n",
       "      <td>0.111351</td>\n",
       "      <td>0.494286</td>\n",
       "      <td>0.487702</td>\n",
       "      <td>0.489176</td>\n",
       "      <td>0.497062</td>\n",
       "      <td>0.497367</td>\n",
       "      <td>0.495841</td>\n",
       "      <td>0.480004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            00001       00004       00005      00006      00007       00008  \\\n",
       "count  100.000000  100.000000  100.000000  100.00000  100.00000  100.000000   \n",
       "mean     0.325000    0.575000    0.535000    0.94500    0.54500    0.415000   \n",
       "std      0.462618    0.494286    0.498761    0.22355    0.49795    0.492648   \n",
       "min      0.000000    0.000000    0.000000    0.00000    0.00000    0.000000   \n",
       "25%      0.000000    0.000000    0.000000    1.00000    0.00000    0.000000   \n",
       "50%      0.000000    1.000000    1.000000    1.00000    1.00000    0.000000   \n",
       "75%      1.000000    1.000000    1.000000    1.00000    1.00000    1.000000   \n",
       "max      1.000000    1.000000    1.000000    1.00000    1.00000    1.000000   \n",
       "\n",
       "           00009       00010       00020       00026       00032       00038  \\\n",
       "count  100.00000  100.000000  100.000000  100.000000  100.000000  100.000000   \n",
       "mean     0.54500    0.985000    0.525000    0.545000    0.410000    0.480000   \n",
       "std      0.49795    0.111351    0.494286    0.487702    0.489176    0.497062   \n",
       "min      0.00000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "25%      0.00000    1.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "50%      1.00000    1.000000    1.000000    1.000000    0.000000    0.000000   \n",
       "75%      1.00000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
       "max      1.00000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
       "\n",
       "            00039       00044       00047  \n",
       "count  100.000000  100.000000  100.000000  \n",
       "mean     0.510000    0.460000    0.370000  \n",
       "std      0.497367    0.495841    0.480004  \n",
       "min      0.000000    0.000000    0.000000  \n",
       "25%      0.000000    0.000000    0.000000  \n",
       "50%      0.750000    0.000000    0.000000  \n",
       "75%      1.000000    1.000000    1.000000  \n",
       "max      1.000000    1.000000    1.000000  "
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "R    54\n",
       "D    44\n",
       "I     2\n",
       "Name: party, dtype: int64"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.party.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Hopkins Statistic**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyclustertend import hopkins\n",
    "from sklearn.preprocessing import scale"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df.drop(['name', 'party', 'state'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 15)"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.16980517450003785"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hopkins(df2, df2.shape[0]) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "H is close to zero, which means that the dataset contains meaningful clusters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13788326303501783"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hopkins(scale(df2), df2.shape[0])  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Scaling caused an improvement in the Hopkins statistic."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**K-Means Clustering - not scaled**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "k_means = KMeans(n_clusters = 2).fit(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "clusters = k_means.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "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>party</th>\n",
       "      <th>D</th>\n",
       "      <th>I</th>\n",
       "      <th>R</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>row_0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "party   D  I   R\n",
       "row_0           \n",
       "0       3  0  54\n",
       "1      41  2   0"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tab = pd.crosstab(clusters, df[\"party\"])\n",
    "tab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "R    54\n",
       "D    44\n",
       "I     2\n",
       "Name: party, dtype: int64"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.party.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Considering the confusion matrix above, it looks like K-Means correctly clustered all the Republicans (they voted according to the Party decisions), but missed 3 Democrat senators.I can check all the row, or write a little code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['predicted_clusters'] = clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "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>name</th>\n",
       "      <th>party</th>\n",
       "      <th>state</th>\n",
       "      <th>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "      <th>predicted_clusters</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>Kirk</td>\n",
       "      <td>R</td>\n",
       "      <td>IL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Collins</td>\n",
       "      <td>R</td>\n",
       "      <td>ME</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Boxer</td>\n",
       "      <td>D</td>\n",
       "      <td>CA</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Cruz</td>\n",
       "      <td>R</td>\n",
       "      <td>TX</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>Shelby</td>\n",
       "      <td>R</td>\n",
       "      <td>AL</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       name party state  00001  00004  00005  00006  00007  00008  00009  \\\n",
       "51     Kirk     R    IL    0.0    1.0    1.0    1.0    1.0    1.0    1.0   \n",
       "20  Collins     R    ME    0.0    1.0    1.0    1.0    1.0    0.0    0.0   \n",
       "9     Boxer     D    CA    0.5    0.0    0.0    1.0    0.0    1.0    0.0   \n",
       "26     Cruz     R    TX    0.0    1.0    1.0    0.0    1.0    0.0    1.0   \n",
       "86   Shelby     R    AL    0.0    1.0    1.0    1.0    1.0    0.0    1.0   \n",
       "\n",
       "    00010  00020  00026  00032  00038  00039  00044  00047  predicted_clusters  \n",
       "51    1.0    1.0    1.0    1.0    0.0    1.0    1.0    0.0                   0  \n",
       "20    1.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0                   0  \n",
       "9     1.0    0.0    0.0    1.0    1.0    0.0    1.0    1.0                   1  \n",
       "26    1.0    1.0    1.0    0.0    0.0    1.0    0.0    0.0                   0  \n",
       "86    1.0    1.0    1.0    0.0    0.0    1.0    0.0    0.0                   0  "
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "D_miss = df[(clusters == 0) & (df[\"party\"] == \"D\")] # 0 is used as a mistake in the Confusion Matrix for Democrats."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "42    Heitkamp\n",
       "56     Manchin\n",
       "74        Reid\n",
       "Name: name, dtype: object"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "D_miss['name']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Elbow Method**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "ssd = []\n",
    "K = range(1,10)\n",
    "for k in K:\n",
    "    kmeans = KMeans(n_clusters = k).fit((df2))\n",
    "    ssd.append(kmeans.inertia_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(K, ssd, \"bx-\")\n",
    "plt.xlabel(\"K\")\n",
    "plt.title(\"Inertia\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[341.335,\n",
       " 61.70297837617299,\n",
       " 48.59646690518784,\n",
       " 38.02198544698544,\n",
       " 32.988652113652115,\n",
       " 28.11730196886447,\n",
       " 24.909895833333334,\n",
       " 21.319334156534957,\n",
       " 18.462257375776396]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ssd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from yellowbrick.cluster import KElbowVisualizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "kmeans = KMeans()\n",
    "visu = KElbowVisualizer(kmeans, k = (1,10))\n",
    "visu.fit(df2)\n",
    "visu.poof();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "kmeans = KMeans()\n",
    "visu = KElbowVisualizer(kmeans, k = (2,10))\n",
    "visu.fit(df2)\n",
    "visu.poof();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The results say that (when the k-search range is between 2 and 10), there are actually 4 clusters; although there are 2 parties and 2 independent senators. Probably, some senators do not always like taking orders while voting. Increasing the k size may give us a feeling about the sub-divisions in the parties (or in just one party). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Increase k to 4 in order to see if there are sub-parties**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans_4 = KMeans(n_clusters = 4).fit(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels_4 = kmeans_4.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2['cluster_4'] = labels_4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "scrolled": true
   },
   "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>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "      <th>cluster_4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   00001  00004  00005  00006  00007  00008  00009  00010  00020  00026  \\\n",
       "0    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    1.0    1.0   \n",
       "1    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    0.0    1.0   \n",
       "2    1.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "3    0.0    1.0    1.0    1.0    1.0    0.0    1.0    1.0    1.0    1.0   \n",
       "4    0.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "\n",
       "   00032  00038  00039  00044  00047  cluster_4  \n",
       "0    0.0    0.0    0.0    0.0    0.0          1  \n",
       "1    0.0    1.0    0.0    1.0    0.0          3  \n",
       "2    1.0    1.0    0.0    1.0    1.0          0  \n",
       "3    0.0    0.0    1.0    0.0    0.0          1  \n",
       "4    0.0    1.0    0.0    1.0    0.0          2  "
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    52\n",
       "0    32\n",
       "3     8\n",
       "2     8\n",
       "Name: cluster_4, dtype: int64"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.cluster_4.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That is an interesting result. It looks like there are sub-parties/groups."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'cluster_4' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-136-d004a295bc0b>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mtab\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcrosstab\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcluster_4\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdf\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m\"party\"\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mtab\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'cluster_4' is not defined"
     ]
    }
   ],
   "source": [
    "tab = pd.crosstab(cluster_4, df[\"party\"])\n",
    "tab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'party'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   2645\u001b[0m             \u001b[1;32mtry\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2646\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2647\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'party'",
      "\nDuring handling of the above exception, another exception occurred:\n",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-107-75672b382087>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mtab_4\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcrosstab\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mclusters\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdf2\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m\"party\"\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mtab_4\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   2798\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mnlevels\u001b[0m \u001b[1;33m>\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2799\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_getitem_multilevel\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2800\u001b[1;33m             \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2801\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mis_integer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2802\u001b[0m                 \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mindexer\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexes\\base.py\u001b[0m in \u001b[0;36mget_loc\u001b[1;34m(self, key, method, tolerance)\u001b[0m\n\u001b[0;32m   2646\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2647\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 2648\u001b[1;33m                 \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_engine\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_loc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_maybe_cast_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   2649\u001b[0m         \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmethod\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mmethod\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtolerance\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mtolerance\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   2650\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mindexer\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mndim\u001b[0m \u001b[1;33m>\u001b[0m \u001b[1;36m1\u001b[0m \u001b[1;32mor\u001b[0m \u001b[0mindexer\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msize\u001b[0m \u001b[1;33m>\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\index.pyx\u001b[0m in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;32mpandas\\_libs\\hashtable_class_helper.pxi\u001b[0m in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[1;34m()\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'party'"
     ]
    }
   ],
   "source": [
    "tab_4 = pd.crosstab(clusters, df2[\"party\"])\n",
    "tab_4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Increase k to 6 in order to see if there are sub-parties**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans_6 = KMeans(n_clusters = 6).fit(df2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels_6 = kmeans_6.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2['cluster_6'] = labels_6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "      <th>predicted_clusters</th>\n",
       "      <th>cluster_id</th>\n",
       "      <th>cluster_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   00001  00004  00005  00006  00007  00008  00009  00010  00020  00026  \\\n",
       "0    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    1.0    1.0   \n",
       "1    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    0.0    1.0   \n",
       "2    1.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "3    0.0    1.0    1.0    1.0    1.0    0.0    1.0    1.0    1.0    1.0   \n",
       "4    0.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "\n",
       "   00032  00038  00039  00044  00047  predicted_clusters  cluster_id  \\\n",
       "0    0.0    0.0    0.0    0.0    0.0                   0           0   \n",
       "1    0.0    1.0    0.0    1.0    0.0                   0           2   \n",
       "2    1.0    1.0    0.0    1.0    1.0                   1           1   \n",
       "3    0.0    0.0    1.0    0.0    0.0                   0           0   \n",
       "4    0.0    1.0    0.0    1.0    0.0                   1           3   \n",
       "\n",
       "   cluster_6  \n",
       "0          0  \n",
       "1          5  \n",
       "2          1  \n",
       "3          0  \n",
       "4          2  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    47\n",
       "1    37\n",
       "2     6\n",
       "4     5\n",
       "3     3\n",
       "5     2\n",
       "Name: cluster_6, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.cluster_6.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "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>cluster_6</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>row_0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>47</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>37</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "cluster_6   0   1  2  3  4  5\n",
       "row_0                        \n",
       "0          47   0  0  3  5  2\n",
       "1           0  37  6  0  0  0"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tab_6 = pd.crosstab(clusters, df2[\"cluster_6\"])\n",
    "tab_6"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "OK. This is the point to stop increasing the number of clusters. k=4 looks like a reasonable result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Silhoutte Score**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import silhouette_samples,silhouette_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Silhouette Score(n=2): 0.7303344476452661\n"
     ]
    }
   ],
   "source": [
    "print(f'Silhouette Score(n=2): {silhouette_score(df2, labels)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Silhouette Score(n=4): 0.8217108978587662\n"
     ]
    }
   ],
   "source": [
    "print(f'Silhouette Score(n=4): {silhouette_score(df2, labels_4)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "0.72 is a very nice Silhoutte score; which may be commented as; clusters are quite dense and separated."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1e519f37188>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(df2.iloc[:,0], df2.iloc[:,1], c = labels, s = 50, cmap = \"viridis\")\n",
    "plt.scatter(kmeans.cluster_centers_[:, 0], kmeans.cluster_centers_[:, 1], s=200, c='black',alpha=0.5, label = 'Centroids')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**ARI Score**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A domain expert can make comments on the relationship between voting to various bills and his/her party, so I will skip the ARI. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**K-Means Clustering - Scaled**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "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>name</th>\n",
       "      <th>party</th>\n",
       "      <th>state</th>\n",
       "      <th>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "      <th>predicted_clusters</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alexander</td>\n",
       "      <td>R</td>\n",
       "      <td>TN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Ayotte</td>\n",
       "      <td>R</td>\n",
       "      <td>NH</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Baldwin</td>\n",
       "      <td>D</td>\n",
       "      <td>WI</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Barrasso</td>\n",
       "      <td>R</td>\n",
       "      <td>WY</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Bennet</td>\n",
       "      <td>D</td>\n",
       "      <td>CO</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        name party state  00001  00004  00005  00006  00007  00008  00009  \\\n",
       "0  Alexander     R    TN    0.0    1.0    1.0    1.0    1.0    0.0    0.0   \n",
       "1     Ayotte     R    NH    0.0    1.0    1.0    1.0    1.0    0.0    0.0   \n",
       "2    Baldwin     D    WI    1.0    0.0    0.0    1.0    0.0    1.0    0.0   \n",
       "3   Barrasso     R    WY    0.0    1.0    1.0    1.0    1.0    0.0    1.0   \n",
       "4     Bennet     D    CO    0.0    0.0    0.0    1.0    0.0    1.0    0.0   \n",
       "\n",
       "   00010  00020  00026  00032  00038  00039  00044  00047  predicted_clusters  \n",
       "0    1.0    1.0    1.0    0.0    0.0    0.0    0.0    0.0                   0  \n",
       "1    1.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0                   0  \n",
       "2    1.0    0.0    0.0    1.0    1.0    0.0    1.0    1.0                   1  \n",
       "3    1.0    1.0    1.0    0.0    0.0    1.0    0.0    0.0                   0  \n",
       "4    1.0    0.0    0.0    0.0    1.0    0.0    1.0    0.0                   1  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "df3 = df.drop(['name', 'party', 'state', 'predicted_clusters'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>00001</th>\n",
       "      <th>00004</th>\n",
       "      <th>00005</th>\n",
       "      <th>00006</th>\n",
       "      <th>00007</th>\n",
       "      <th>00008</th>\n",
       "      <th>00009</th>\n",
       "      <th>00010</th>\n",
       "      <th>00020</th>\n",
       "      <th>00026</th>\n",
       "      <th>00032</th>\n",
       "      <th>00038</th>\n",
       "      <th>00039</th>\n",
       "      <th>00044</th>\n",
       "      <th>00047</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   00001  00004  00005  00006  00007  00008  00009  00010  00020  00026  \\\n",
       "0    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    1.0    1.0   \n",
       "1    0.0    1.0    1.0    1.0    1.0    0.0    0.0    1.0    0.0    1.0   \n",
       "2    1.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "3    0.0    1.0    1.0    1.0    1.0    0.0    1.0    1.0    1.0    1.0   \n",
       "4    0.0    0.0    0.0    1.0    0.0    1.0    0.0    1.0    0.0    0.0   \n",
       "\n",
       "   00032  00038  00039  00044  00047  \n",
       "0    0.0    0.0    0.0    0.0    0.0  \n",
       "1    0.0    1.0    0.0    1.0    0.0  \n",
       "2    1.0    1.0    0.0    1.0    1.0  \n",
       "3    0.0    0.0    1.0    0.0    0.0  \n",
       "4    0.0    1.0    0.0    1.0    0.0  "
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 15)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13179016969095544"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hopkins(scale(df3), df3.shape[0])  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "k_means_s = KMeans(n_clusters = 2).fit(scale(df3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "clusters_s = k_means_s.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "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>party</th>\n",
       "      <th>D</th>\n",
       "      <th>I</th>\n",
       "      <th>R</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>row_0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>41</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "party   D  I   R\n",
       "row_0           \n",
       "0       3  0  54\n",
       "1      41  2   0"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tab = pd.crosstab(clusters_s, df[\"party\"])\n",
    "tab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We got the same results. Scaling did not cause a huge improvement here."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Elbow Method for the Scaled Set**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "ssd = []\n",
    "K = range(1,10)\n",
    "for k in K:\n",
    "    kmeans = KMeans(n_clusters = k).fit((df3))\n",
    "    ssd.append(kmeans.inertia_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(K, ssd, \"bx-\")\n",
    "plt.xlabel(\"K\")\n",
    "plt.title(\"Inertia\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "kmeans = KMeans()\n",
    "visu = KElbowVisualizer(kmeans, k = (1,10))\n",
    "visu.fit(df2)\n",
    "visu.poof();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yellow brick still says 3 clusters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}