clustering_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Customer segmentation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You are the owner of a shop. It doesn't matter if you own an e-commerce or a  supermarket. It doesn't matter if it is a small shop or a huge company such as Amazon or Netflix, it's better to know your customers.\n",
    "\n",
    "You were able to collect basic data about your customers holding a membership card such as Customer ID, age, gender, annual income, and spending score. This last one is a score based on customer behavior and purchasing data.\n",
    "There are some new products on the market that you are interested in selling. But you want to target a specific type of clients for each one of the products.  \n",
    "\n",
    "Machine learning comes in handy for this task. Particularly, clustering, the most important unsupervised learning problem, is able to create categories grouping similar individuals.\n",
    "These categories are called clusters. A cluster is a collection of points in a dataset. These points are more similar between them than they are to points belonging to other clusters.\n",
    "Distance-based clustering groups the points into some number of clusters such that distances within the cluster should be small while distances between clusters should be large."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import modules requiered"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First of all, we need to import the required module. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For this particular project, we'll work with two `scikit-learn` modules: `Kmeans` and `PCA`. They will allow us to perform a clustering algorithm and dimensionality reduction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import MinMaxScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from plotly.offline import iplot, init_notebook_mode\n",
    "import plotly.graph_objs as go\n",
    "import plotly.io as pio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import extra_graphs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read data into a DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We read the basic data stored in the [customers.csv](https://www.kaggle.com/akram24/mall-customers) file into a `DataFrame` using pandas. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "customers = pd.read_csv(\"customers.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We check the first five rows of the DataFrame. We can see that we have: CustumerID, Gender, Age, Annual Income expressed as price x1000, and the spending score as we expected."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>CustomerID</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Female</td>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Female</td>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerID  Gender  Age  Annual Income (k$)  Spending Score (1-100)\n",
       "0           1    Male   19                  15                      39\n",
       "1           2    Male   21                  15                      81\n",
       "2           3  Female   20                  16                       6\n",
       "3           4  Female   23                  16                      77\n",
       "4           5  Female   31                  17                      40"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "customers.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exploring the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, it's time to explore the data to check the quality of the data and the distribution of the variables."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we check that if there is any missing value in the dataset. K-means algorithm is not able to deal with missing values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missing values in each variable: \n",
      "CustomerID                0\n",
      "Gender                    0\n",
      "Age                       0\n",
      "Annual Income (k$)        0\n",
      "Spending Score (1-100)    0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(f\"Missing values in each variable: \\n{customers.isnull().sum()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fortunately, there is no missing data. We can also check if there are duplicated rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duplicated rows: 0\n"
     ]
    }
   ],
   "source": [
    "print(f\"Duplicated rows: {customers.duplicated().sum()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we check how each variable is presented in the DataFrame. Categorical variables cannot be handled directly. K-means is based on distances. The approach for converting those variables depend on the type of categorical variables. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variable:                  Type: \n",
      "CustomerID                 int64\n",
      "Gender                    object\n",
      "Age                        int64\n",
      "Annual Income (k$)         int64\n",
      "Spending Score (1-100)     int64\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "print(f\"Variable:                  Type: \\n{customers.dtypes}\") "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After that, we can start observing the distribution of the variables. Here, we'll define two functions. The first one will retrieve descriptive statistics of the variables. The second one will help us graph the variable distribution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Descriptive statistics and Distribution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the descriptive statistcs, we'll get mean, standard deviation, median and variance. If the variable is not numeric, we'll get the counts in each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def statistics(variable):\n",
    "    if variable.dtype == \"int64\" or variable.dtype == \"float64\":\n",
    "        return pd.DataFrame([[variable.name, np.mean(variable), np.std(variable), np.median(variable), np.var(variable)]], \n",
    "                            columns = [\"Variable\", \"Mean\", \"Standard Deviation\", \"Median\", \"Variance\"]).set_index(\"Variable\")\n",
    "    else:\n",
    "        return pd.DataFrame(variable.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def graph_histo(x):\n",
    "    if x.dtype == \"int64\" or x.dtype == \"float64\":\n",
    "        # Select size of bins by getting maximum and minimum and divide the substraction by 10\n",
    "        size_bins = 10\n",
    "        # Get the title by getting the name of the column\n",
    "        title = x.name\n",
    "        #Assign random colors to each graph\n",
    "        color_kde = list(map(float, np.random.rand(3,)))\n",
    "        color_bar = list(map(float, np.random.rand(3,)))\n",
    "\n",
    "        # Plot the displot\n",
    "        sns.distplot(x, bins=size_bins, kde_kws={\"lw\": 1.5, \"alpha\":0.8, \"color\":color_kde},\n",
    "                       hist_kws={\"linewidth\": 1.5, \"edgecolor\": \"grey\",\n",
    "                                \"alpha\": 0.4, \"color\":color_bar})\n",
    "        # Customize ticks and labels\n",
    "        plt.xticks(size=14)\n",
    "        plt.yticks(size=14);\n",
    "        plt.ylabel(\"Frequency\", size=16, labelpad=15);\n",
    "        # Customize title\n",
    "        plt.title(title, size=18)\n",
    "        # Customize grid and axes visibility\n",
    "        plt.grid(False);\n",
    "        plt.gca().spines[\"top\"].set_visible(False);\n",
    "        plt.gca().spines[\"right\"].set_visible(False);\n",
    "        plt.gca().spines[\"bottom\"].set_visible(False);\n",
    "        plt.gca().spines[\"left\"].set_visible(False);   \n",
    "    else:\n",
    "        x = pd.DataFrame(x)\n",
    "        # Plot       \n",
    "        sns.catplot(x=x.columns[0], kind=\"count\", palette=\"spring\", data=x)\n",
    "        # Customize title\n",
    "        title = x.columns[0]\n",
    "        plt.title(title, size=18)\n",
    "        # Customize ticks and labels\n",
    "        plt.xticks(size=14)\n",
    "        plt.yticks(size=14);\n",
    "        plt.xlabel(\"\")\n",
    "        plt.ylabel(\"Counts\", size=16, labelpad=15);        \n",
    "        # Customize grid and axes visibility\n",
    "        plt.gca().spines[\"top\"].set_visible(False);\n",
    "        plt.gca().spines[\"right\"].set_visible(False);\n",
    "        plt.gca().spines[\"bottom\"].set_visible(False);\n",
    "        plt.gca().spines[\"left\"].set_visible(False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll start by the **Spending Score**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "spending = customers[\"Spending Score (1-100)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Mean</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Median</th>\n",
       "      <th>Variance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variable</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <td>50.2</td>\n",
       "      <td>25.758882</td>\n",
       "      <td>50.0</td>\n",
       "      <td>663.52</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        Mean  Standard Deviation  Median  Variance\n",
       "Variable                                                          \n",
       "Spending Score (1-100)  50.2           25.758882    50.0    663.52"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "statistics(spending)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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": [
    "graph_histo(spending)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, we'll check **Age**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "age = customers[\"Age\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Mean</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Median</th>\n",
       "      <th>Variance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variable</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>38.85</td>\n",
       "      <td>13.934041</td>\n",
       "      <td>36.0</td>\n",
       "      <td>194.1575</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Mean  Standard Deviation  Median  Variance\n",
       "Variable                                             \n",
       "Age       38.85           13.934041    36.0  194.1575"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "statistics(age)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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": [
    "graph_histo(age)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we'll explore **Annual Income** variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "income = customers[\"Annual Income (k$)\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Mean</th>\n",
       "      <th>Standard Deviation</th>\n",
       "      <th>Median</th>\n",
       "      <th>Variance</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variable</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <td>60.56</td>\n",
       "      <td>26.198977</td>\n",
       "      <td>61.5</td>\n",
       "      <td>686.3864</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Mean  Standard Deviation  Median  Variance\n",
       "Variable                                                       \n",
       "Annual Income (k$)  60.56           26.198977    61.5  686.3864"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "statistics(income)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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": [
    "graph_histo(income)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "gender = customers[\"Gender\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Female</th>\n",
       "      <td>112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Male</th>\n",
       "      <td>88</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Gender\n",
       "Female     112\n",
       "Male        88"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "statistics(gender)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "graph_histo(gender)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Correlation between parameteres"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also, we will analyze the correlation between the numeric parameters. For that aim, we'll use the `pairplot` seaborn function. We want to see whether there is a difference between gender. So, we are going to set the `hue` parameter to get different colors for points belonging to female or customers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504.625x432 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(customers, x_vars = [\"Age\", \"Annual Income (k$)\", \"Spending Score (1-100)\"], \n",
    "               y_vars = [\"Age\", \"Annual Income (k$)\", \"Spending Score (1-100)\"], \n",
    "               hue = \"Gender\", \n",
    "               kind= \"scatter\",\n",
    "               palette = \"YlGnBu\",\n",
    "               height = 2,\n",
    "               plot_kws={\"s\": 35, \"alpha\": 0.8});"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Why is it important to look into the descriptive statistics, distribution and correlation between variables?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to apply K-means, we need to meet the algorithm assumptions. \n",
    "  \n",
    "K-means assumes:\n",
    "\n",
    "- **Cluster's shape**: The variance of the distribution is spherical meaning that clusters have a spherical shape. In order for this to be true, all variables should be normally distributed and have the same variance.\n",
    "- **Clusters' Size**: All clusters have the same number of observations.\n",
    "- **Relationship between variables**: There is little or no correlation between the variables."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our dataset, our variables are normally distributed. Variances are quite close to each other. Except for age that has a lower variance that the rest of the variables.\n",
    "We could find a proper transformation to solve this issue. We could apply the logarithm or Box-Cox transformation. \n",
    "Box-Cox is a family of transformations which allows us to correct non-normal distributed variables or non-equal variances."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dimensionality reduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After we checked that we can apply k-means, we can apply Principal Component Analysis (PCA) to discover which dimensions best maximize the variance of features involved."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Principal Component Analysis (PCA)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we'll transform the categorical variable into two binary variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "customers[\"Male\"] = customers.Gender.apply(lambda x: 0 if x == \"Male\" else 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "customers[\"Female\"] = customers.Gender.apply(lambda x: 0 if x == \"Female\" else 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, we are going to select from the dataset all the useful columns. Customer ID is not a useful feature. Gender will split it into two binaries categories. It should not appear in the final dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = customers.iloc[:, 2:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "      <th>Male</th>\n",
       "      <th>Female</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>39</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>81</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Annual Income (k$)  Spending Score (1-100)  Male  Female\n",
       "0   19                  15                      39     0       1\n",
       "1   21                  15                      81     0       1\n",
       "2   20                  16                       6     1       0\n",
       "3   23                  16                      77     1       0\n",
       "4   31                  17                      40     1       0"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to apply PCA, we are going to use the `PCA` function from sklearn module."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Apply PCA and fit the features selected\n",
    "pca = PCA(n_components=2).fit(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "During the fitting process, the model learns some quantities from the data: the \"components\" and \"explained variance\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.88980385e-01  5.88604475e-01  7.86022241e-01  3.32880772e-04\n",
      "  -3.32880772e-04]\n",
      " [ 1.30957602e-01  8.08400899e-01 -5.73875514e-01 -1.57927017e-03\n",
      "   1.57927017e-03]]\n"
     ]
    }
   ],
   "source": [
    "print(pca.components_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[700.26450987 684.33354753]\n"
     ]
    }
   ],
   "source": [
    "print(pca.explained_variance_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These numbers that appear to be abstract define vectors. The components define the direction of the vector while the explained variance define the squared-length of the vector.\n",
    "\n",
    "The vectors represent the principal axes of the data. The length of the vector indicates the importance of that axis in describing the distribution of the data. The projection of each data point onto the principal axes are the principal components of the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Transform samples using the PCA fit\n",
    "pca_2d = pca.transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can represent this using a type of scatter plot called biplot. Each point is represented by its score regarding the principal components.\n",
    "It is helpful to understand the reduced dimensions of the data. It also helps us discover relationships between the principal components and the original variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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": [
    "extra_graphs.biplot(pca_2d[:,0:2], np.transpose(pca.components_[0:2, :]), labels=X.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can observe that Annual Income as well as Spending Score at the two most important components."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### K-means clustering "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Time for clustering!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to cluster data, we need to determine how to tell if two data points are similar. A proximity measure characterizes the similarity or dissimilarity that exists between objects.\n",
    "  \n",
    "We can choose to determine if two points are similar. So if the value is large, the points are very similar.\n",
    "Or choose to determine if they are dissimilar. If the value is small, the points are similar. This is what we know as \"distance\".\n",
    "  \n",
    "There are various distances that a clustering algorithm can use: Manhattan distance, Minkowski distance, Euclidean distance, among others."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "${\\sqrt{\\sum_{i=1}^n (x_i-y_i)^2}}$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "K-means typically uses Euclidean distance to determine how similar (or dissimilar) two points are."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we need to fix the numbers of clusters to use. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several direct methods to perform this. Among them, we find the elbow and silhouette methods."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll consider the total intra-cluster variation (or total within-cluster sum of square (WSS)). The goal is to minimize WSS."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Elbow method looks at how the total WSS varies with the number of clusters. \n",
    "For that, we'll compute k-means for a range of different values of k. Then, we calculate the total WSS. We plot the curve WSS vs. number of clusters. \n",
    "Finally, we locate the elbow or bend of the plot. This point is considered to be the appropriate number of clusters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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": [
    "wcss = []\n",
    "for i in range(1,11):\n",
    "    km = KMeans(n_clusters=i,init='k-means++', max_iter=300, n_init=10, random_state=0)\n",
    "    km.fit(X)\n",
    "    wcss.append(km.inertia_)\n",
    "plt.plot(range(1,11),wcss, c=\"#c51b7d\")\n",
    "plt.gca().spines[\"top\"].set_visible(False)\n",
    "plt.gca().spines[\"right\"].set_visible(False)\n",
    "plt.title('Elbow Method', size=14)\n",
    "plt.xlabel('Number of clusters', size=12)\n",
    "plt.ylabel('wcss', size=14)\n",
    "plt.show() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How does k-means clustering works? The main idea is to select k centers, one for each cluster. There are several ways to initialize those centers. We can do it randomly, pass certain points that we believe are the center or place them in a smart way (e.g. as far away from each other as possible).\n",
    "Then, we calculate the Euclidean distance between each point and the cluster centers. We assign the points to the cluster center where the distance is minimum.\n",
    "After that, we recalculate the new cluster center. We select the point that is in the middle of each cluster as the new center. \n",
    "And we start again, calculate distance, assign to cluster, calculate new centers. When do we stop? When the centers do not move anymore."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Kmeans algorithm\n",
    "# n_clusters: Number of clusters. In our case 5\n",
    "# init: k-means++. Smart initialization\n",
    "# max_iter: Maximum number of iterations of the k-means algorithm for a single run\n",
    "# n_init: Number of time the k-means algorithm will be run with different centroid seeds. \n",
    "# random_state: Determines random number generation for centroid initialization.\n",
    "kmeans = KMeans(n_clusters=5, init='k-means++', max_iter=10, n_init=10, random_state=0)\n",
    "\n",
    "# Fit and predict \n",
    "y_means = kmeans.fit_predict(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's check how our clusters look like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize = (8, 6))\n",
    "\n",
    "plt.scatter(pca_2d[:, 0], pca_2d[:, 1],\n",
    "            c=y_means, \n",
    "            edgecolor=\"none\", \n",
    "            cmap=plt.cm.get_cmap(\"Spectral_r\", 5),\n",
    "            alpha=0.5)\n",
    "        \n",
    "plt.gca().spines[\"top\"].set_visible(False)\n",
    "plt.gca().spines[\"right\"].set_visible(False)\n",
    "plt.gca().spines[\"bottom\"].set_visible(False)\n",
    "plt.gca().spines[\"left\"].set_visible(False)\n",
    "\n",
    "plt.xticks(size=12)\n",
    "plt.yticks(size=12)\n",
    "\n",
    "plt.xlabel(\"Component 1\", size = 14, labelpad=10)\n",
    "plt.ylabel(\"Component 2\", size = 14, labelpad=10)\n",
    "\n",
    "plt.title('Dominios agrupados en 5 clusters', size=16)\n",
    "\n",
    "\n",
    "plt.colorbar(ticks=[0, 1, 2, 3, 4]);\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "centroids = pd.DataFrame(kmeans.cluster_centers_, columns = [\"Age\", \"Annual Income\", \"Spending\", \"Male\", \"Female\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "centroids.index_name = \"ClusterID\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "centroids[\"ClusterID\"] = centroids.index\n",
    "centroids = centroids.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "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>Age</th>\n",
       "      <th>Annual Income</th>\n",
       "      <th>Spending</th>\n",
       "      <th>Male</th>\n",
       "      <th>Female</th>\n",
       "      <th>ClusterID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>45.217391</td>\n",
       "      <td>26.304348</td>\n",
       "      <td>20.913043</td>\n",
       "      <td>0.608696</td>\n",
       "      <td>0.391304</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>32.692308</td>\n",
       "      <td>86.538462</td>\n",
       "      <td>82.128205</td>\n",
       "      <td>0.538462</td>\n",
       "      <td>0.461538</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>43.088608</td>\n",
       "      <td>55.291139</td>\n",
       "      <td>49.569620</td>\n",
       "      <td>0.582278</td>\n",
       "      <td>0.417722</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>40.666667</td>\n",
       "      <td>87.750000</td>\n",
       "      <td>17.583333</td>\n",
       "      <td>0.472222</td>\n",
       "      <td>0.527778</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25.521739</td>\n",
       "      <td>26.304348</td>\n",
       "      <td>78.565217</td>\n",
       "      <td>0.608696</td>\n",
       "      <td>0.391304</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Age  Annual Income   Spending      Male    Female  ClusterID\n",
       "0  45.217391      26.304348  20.913043  0.608696  0.391304          0\n",
       "1  32.692308      86.538462  82.128205  0.538462  0.461538          1\n",
       "2  43.088608      55.291139  49.569620  0.582278  0.417722          2\n",
       "3  40.666667      87.750000  17.583333  0.472222  0.527778          3\n",
       "4  25.521739      26.304348  78.565217  0.608696  0.391304          4"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "centroids"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The most important features appear to be Annual Income and Spending score. \n",
    "We have people whose income is low but spend in the same range - segment 0. People whose earnings a high and spend a lot - segment 1. Customers whose income is middle range but also spend at the same level - segment 2. \n",
    "Then we have customers whose income is very high but they have most spendings - segment 4. And last, people whose earnings are little but they spend a lot- segment 5."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Imagine that tomorrow we have a new member. And we want to know which segment that person belongs. We can predict this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The new customer belongs to segment 2\n"
     ]
    }
   ],
   "source": [
    "X_new = np.array([[43, 76, 56, 0, 1]]) \n",
    " \n",
    "new_customer = kmeans.predict(X_new)\n",
    "print(f\"The new customer belongs to segment {new_customer[0]}\")"
   ]
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}