{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "##**Visualizing and Removing Outliers Using Box Plots**"
      ],
      "metadata": {
        "id": "PPfE1YVl2CE1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import sklearn\n",
        "from sklearn.datasets import load_diabetes\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "diabetes = load_diabetes()\n",
        "\n",
        "column_name = diabetes.feature_names\n",
        "df_diabetics = pd.DataFrame(diabetes.data, columns=column_name)\n",
        "\n",
        "sns.boxplot(df_diabetics['bmi'])\n",
        "plt.title('Boxplot of BMI')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 433
        },
        "id": "MDJ1U7b21A6w",
        "outputId": "7635ee6e-239c-44c1-bb09-50e99e6935df"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def removal_box_plot(df, column, threshold):\n",
        "    removed_outliers = df[df[column] <= threshold]\n",
        "\n",
        "    sns.boxplot(removed_outliers[column])\n",
        "    plt.title(f'Box Plot without Outliers of {column}')\n",
        "    plt.show()\n",
        "    return removed_outliers\n",
        "\n",
        "threshold_value = 0.12\n",
        "\n",
        "no_outliers = removal_box_plot(df_diabetics, 'bmi', threshold_value)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 436
        },
        "id": "orziZE2x1A9V",
        "outputId": "8d91f68c-2a4e-4e4a-c22f-f9fae7e15ebb"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##**Visualizing and Removing Outliers Using Scatter Plots**"
      ],
      "metadata": {
        "id": "M8gYlEKR4TSB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(6, 4))\n",
        "ax.scatter(df_diabetics['bmi'], df_diabetics['bp'])\n",
        "ax.set_xlabel('BMI')\n",
        "ax.set_ylabel('Blood Pressure')\n",
        "plt.title('Scatter Plot of BMI vs Blood Pressure')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 410
        },
        "id": "AAE1tAw31A_n",
        "outputId": "9d657ea2-df97-4dfc-d0c2-32ab6372b637"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "outlier_indices = np.where((df_diabetics['bmi'] > 0.12) & (df_diabetics['bp'] < 0.8))\n",
        "\n",
        "no_outliers = df_diabetics.drop(outlier_indices[0])\n",
        "\n",
        "fig, ax_no_outliers = plt.subplots(figsize=(6, 4))\n",
        "ax_no_outliers.scatter(no_outliers['bmi'], no_outliers['bp'])\n",
        "ax_no_outliers.set_xlabel('(body mass index of people)')\n",
        "ax_no_outliers.set_ylabel('(bp of the people )')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        },
        "id": "ks0rLBT_1BB-",
        "outputId": "1a5ad8f9-0e04-4275-996d-d9f6a779eedc"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##**Z-Score Method for Outlier Detection**"
      ],
      "metadata": {
        "id": "8BlIrGV76lzJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from scipy import stats\n",
        "import numpy as np\n",
        "z = np.abs(stats.zscore(df_diabetics['age']))\n",
        "print(z)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "_dIeoT8m1BEc",
        "outputId": "d5f004f9-46fb-4e43-acac-47ac35840d89"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.80050009 0.03956713 1.79330681 1.87244107 0.11317236 1.94881082\n",
            " 0.9560041  1.33508832 0.87686984 1.49059233 2.02518057 0.57139085\n",
            " 0.34228161 0.11317236 0.95323959 1.1087436  0.11593688 1.48782782\n",
            " 0.80326461 0.57415536 1.03237385 1.79607132 1.79607132 0.95323959\n",
            " 1.33785284 1.41422259 2.25428981 0.49778562 1.10597908 1.41145807\n",
            " 1.26148309 0.49778562 0.72413034 0.6477606  0.34228161 1.02960933\n",
            " 0.26591186 0.19230663 0.03956713 0.03956713 0.11317236 2.10155031\n",
            " 1.26148309 0.41865135 0.95323959 0.57139085 1.18511334 1.64333183\n",
            " 1.41145807 0.87963435 0.72413034 1.25871858 1.1087436  0.19230663\n",
            " 1.03237385 0.87963435 0.87963435 0.57415536 0.87686984 1.33508832\n",
            " 1.49059233 0.87963435 0.57415536 0.72689486 1.41145807 0.9560041\n",
            " 0.19230663 0.87686984 0.80050009 0.34228161 0.03956713 0.03956713\n",
            " 1.33508832 0.26591186 0.26591186 0.19230663 0.65052511 2.02518057\n",
            " 0.11317236 2.17792006 1.48782782 0.26591186 0.34504612 0.80326461\n",
            " 0.03680262 0.95323959 1.49059233 0.95323959 1.1087436  0.9560041\n",
            " 0.26591186 0.95323959 0.42141587 1.03237385 1.64333183 1.49059233\n",
            " 1.18234883 0.57415536 0.03680262 0.03956713 0.34228161 0.34228161\n",
            " 1.94881082 1.25871858 0.57415536 0.4950211  2.02518057 0.57139085\n",
            " 0.41865135 0.80050009 0.87686984 0.41865135 1.79607132 0.41865135\n",
            " 0.4950211  0.65052511 1.02960933 1.25871858 1.18511334 0.34228161\n",
            " 1.03237385 1.33508832 1.02960933 0.11317236 0.11593688 0.11593688\n",
            " 1.87244107 0.72413034 1.1087436  0.18954211 1.33785284 2.02518057\n",
            " 0.34228161 0.87963435 1.56696208 0.11593688 1.94881082 0.11317236\n",
            " 0.72413034 0.4950211  0.87686984 0.57415536 0.87686984 0.65052511\n",
            " 0.6477606  0.87963435 0.65052511 1.18511334 1.26148309 1.03237385\n",
            " 0.4950211  0.03680262 0.72689486 0.87686984 1.41145807 0.57415536\n",
            " 0.34504612 0.03956713 0.26867637 0.11593688 0.19230663 0.9560041\n",
            " 1.1087436  0.34228161 0.95323959 0.87963435 1.18511334 1.48782782\n",
            " 0.03680262 0.03956713 0.4950211  0.42141587 0.87686984 1.33785284\n",
            " 0.34228161 1.41145807 0.95323959 1.02960933 0.87686984 0.49778562\n",
            " 0.80326461 1.02960933 0.95323959 0.95323959 0.34228161 1.56696208\n",
            " 1.71970158 1.41422259 0.11317236 0.03956713 0.18954211 0.11593688\n",
            " 1.18234883 0.18954211 1.41422259 0.57139085 0.49778562 1.02960933\n",
            " 1.1087436  0.87686984 1.18234883 0.72689486 1.71693706 0.03956713\n",
            " 2.32789504 0.65052511 0.03680262 0.18954211 0.6477606  0.80050009\n",
            " 0.18954211 1.9460463  1.41145807 0.03680262 0.6477606  0.57139085\n",
            " 0.26591186 1.56419757 0.87963435 1.87244107 0.4950211  0.9560041\n",
            " 0.49778562 2.10155031 0.57415536 0.6477606  2.17792006 1.41145807\n",
            " 1.1087436  0.57415536 0.80326461 0.18954211 0.26591186 1.41145807\n",
            " 0.95323959 1.41145807 0.57139085 1.18234883 0.72413034 0.4950211\n",
            " 1.02960933 0.6477606  2.17792006 0.34228161 1.26148309 0.57415536\n",
            " 0.87686984 1.71970158 0.87963435 0.26867637 1.41145807 1.1087436\n",
            " 0.11317236 1.71693706 0.6477606  0.03680262 1.03237385 0.57415536\n",
            " 1.64056731 0.26591186 0.87686984 1.02960933 0.34504612 1.56696208\n",
            " 0.72413034 0.72689486 1.1087436  1.25871858 1.33508832 0.18954211\n",
            " 0.11317236 0.80050009 0.26591186 1.56419757 0.34228161 0.11593688\n",
            " 0.26591186 0.72689486 1.41145807 0.80050009 0.18954211 1.94881082\n",
            " 1.48782782 0.34504612 0.87686984 0.26591186 0.80326461 0.95323959\n",
            " 1.48782782 1.56696208 1.25871858 1.56419757 0.18954211 1.49059233\n",
            " 0.4950211  1.1087436  1.41145807 0.03680262 0.4950211  0.80050009\n",
            " 0.34228161 0.03956713 0.26591186 1.56419757 0.87686984 0.19230663\n",
            " 0.18954211 1.41145807 0.03680262 0.19230663 0.11593688 2.02241605\n",
            " 1.56696208 1.25871858 0.49778562 0.18954211 0.34228161 0.41865135\n",
            " 1.86967656 0.41865135 0.49778562 2.02241605 0.4950211  1.48782782\n",
            " 0.6477606  0.03956713 0.95323959 1.56419757 0.80326461 0.26867637\n",
            " 0.18954211 1.71693706 0.6477606  0.57139085 1.26148309 0.11317236\n",
            " 0.42141587 0.41865135 1.33785284 0.57139085 0.34504612 0.6477606\n",
            " 1.18234883 0.42141587 2.25428981 1.71693706 0.11317236 0.80050009\n",
            " 0.6477606  0.03680262 0.57415536 1.79607132 0.26591186 1.1087436\n",
            " 0.49778562 1.56696208 0.11593688 1.26148309 0.42141587 0.80050009\n",
            " 0.34228161 0.87686984 0.41865135 1.03237385 0.03680262 0.72413034\n",
            " 0.9560041  0.19230663 0.34504612 0.19230663 0.41865135 1.10597908\n",
            " 0.57415536 1.56696208 2.25428981 0.95323959 0.03956713 0.41865135\n",
            " 0.34228161 0.03956713 0.34228161 1.49059233 1.02960933 0.11317236\n",
            " 0.72413034 0.4950211  0.41865135 0.9560041  1.10597908 0.11593688\n",
            " 0.18954211 0.49778562 0.87963435 1.56696208 0.72413034 1.26148309\n",
            " 1.79607132 1.10597908 0.26591186 1.25871858 0.49778562 0.34228161\n",
            " 2.32789504 0.42141587 0.34504612 1.02960933 1.18511334 0.57139085\n",
            " 1.33508832 1.1087436  0.19230663 0.11317236 1.56419757 1.1087436\n",
            " 1.71693706 0.11593688 0.57415536 1.1087436  0.18954211 0.42141587\n",
            " 0.4950211  0.80050009 1.64333183 0.18954211 0.03680262 1.64333183\n",
            " 0.6477606  0.72689486 1.02960933 0.87963435 0.19230663 1.48782782\n",
            " 0.18954211 0.57415536 0.34228161 0.26867637 1.18511334 0.87686984\n",
            " 0.11593688 0.87686984 0.9560041  0.9560041 ]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Trimming**"
      ],
      "metadata": {
        "id": "WJQDMKdj94Vz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "threshold_z = 2\n",
        "\n",
        "outlier_indices = np.where(z > threshold_z)[0]\n",
        "no_outliers = df_diabetics.drop(outlier_indices)\n",
        "\n",
        "print(\"Original DataFrame Shape:\", df_diabetics.shape)\n",
        "print(\"DataFrame Shape after Removing Outliers:\", no_outliers.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "apx7-5V91BGn",
        "outputId": "99283adc-a04d-40c4-d9fa-c7dd3d7d3565"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original DataFrame Shape: (442, 10)\n",
            "DataFrame Shape after Removing Outliers: (426, 10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **capping**"
      ],
      "metadata": {
        "id": "pK2GWtxK-q7Y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "threshold_z = 2\n",
        "\n",
        "df_capped = df_diabetics.copy()\n",
        "\n",
        "df_capped['age'] = np.where(z > threshold_z,\n",
        "                            df_diabetics['age'].mean() + threshold_z * df_diabetics['age'].std(),\n",
        "                            df_diabetics['age'])\n",
        "\n",
        "print(\"Original DataFrame Shape:\", df_diabetics.shape)\n",
        "print(\"DataFrame Shape after Capping Outliers:\", df_capped.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G5tgTNas1BJF",
        "outputId": "1181549b-d244-427b-e5a9-45c68c97ece6"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original DataFrame Shape: (442, 10)\n",
            "DataFrame Shape after Capping Outliers: (442, 10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Interquartile Range (IQR) Method**"
      ],
      "metadata": {
        "id": "KyyVY84CAX1u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "Q1 = np.percentile(df_diabetics['bmi'], 25, method='midpoint')\n",
        "Q3 = np.percentile(df_diabetics['bmi'], 75, method='midpoint')\n",
        "IQR = Q3 - Q1\n",
        "print(IQR)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bgC-cNA8-p_Q",
        "outputId": "8bdcbfe5-57b0-453d-f32d-25be5532a662"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.06520763046978838\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "upper = Q3+1.5*IQR\n",
        "upper_array = np.array(df_diabetics['bmi'] >= upper)\n",
        "print(\"Upper Bound:\", upper)\n",
        "print(upper_array.sum())\n",
        "\n",
        "lower = Q1-1.5*IQR\n",
        "lower_array = np.array(df_diabetics['bmi'] <= lower)\n",
        "print(\"Lower Bound:\", lower)\n",
        "print(lower_array.sum())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_fjw3moI-qDG",
        "outputId": "0ed6febf-cf32-4c97-b7ee-654de762ace6"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Upper Bound: 0.12879000811776306\n",
            "3\n",
            "Lower Bound: -0.13204051376139045\n",
            "0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Trimming**"
      ],
      "metadata": {
        "id": "-OyVj7OMCvAx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import sklearn\n",
        "from sklearn.datasets import load_diabetes\n",
        "import pandas as pd\n",
        "\n",
        "diabetes = load_diabetes()\n",
        "\n",
        "column_name = diabetes.feature_names\n",
        "df_diabetes = pd.DataFrame(diabetes.data)\n",
        "df_diabetes.columns = column_name\n",
        "\n",
        "print(\"Old Shape:\", df_diabetes.shape)\n",
        "\n",
        "Q1 = df_diabetes['bmi'].quantile(0.25)\n",
        "Q3 = df_diabetes['bmi'].quantile(0.75)\n",
        "IQR = Q3 - Q1\n",
        "\n",
        "lower = Q1 - 1.5 * IQR\n",
        "upper = Q3 + 1.5 * IQR\n",
        "\n",
        "upper_array = np.where(df_diabetes['bmi'] >= upper)[0]\n",
        "lower_array = np.where(df_diabetes['bmi'] <= lower)[0]\n",
        "\n",
        "df_diabetes.drop(index=upper_array, inplace=True)\n",
        "df_diabetes.drop(index=lower_array, inplace=True)\n",
        "\n",
        "print(\"New Shape:\", df_diabetes.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mczBnyrkCty7",
        "outputId": "82b8415f-e70c-4760-bd0c-51f1d4a3b14c"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Old Shape: (442, 10)\n",
            "New Shape: (439, 10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **capping**"
      ],
      "metadata": {
        "id": "1mAfq6YADVVB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_capped = df_diabetes.copy()\n",
        "\n",
        "df_capped['bmi'] = np.where(df_capped['bmi'] > upper, upper, df_capped['bmi'])\n",
        "df_capped['bmi'] = np.where(df_capped['bmi'] < lower, lower, df_capped['bmi'])\n",
        "\n",
        "print(\"Shape after Capping:\", df_capped.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0p-vbRdC-qGq",
        "outputId": "cc4532a6-c20e-4a53-b589-a50f8a8b0404"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Shape after Capping: (439, 10)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "fM1k73e--qK5"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}