notebooks/CH14_Github.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# 챕터 14 실습 코드들 \n",
        "\n",
        "챕터 14의 실습에 필요한 코드 예제를 구글 코랩에서 여시려면 [링크](https://colab.research.google.com/drive/1M4gn0eDmkmoBJ6rG6HF_WnTjMWIuUXxA?usp=sharing)를 클릭하시거나 아래의 주소를 웹브라우저에 복사해 주세요. \n",
        "\n",
        "실습 코드 링크: https://colab.research.google.com/drive/1M4gn0eDmkmoBJ6rG6HF_WnTjMWIuUXxA?usp=sharing"
      ],
      "metadata": {
        "id": "SU8muODN0X4c"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-1: 실습에 필요한 라이브러리들]"
      ],
      "metadata": {
        "id": "CsUbasblHp8G"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CeyFB5MGQP1q"
      },
      "source": [
        "# 필요한 라이브러리 로딩\n",
        "import numpy as np\n",
        "import pandas as pd \n",
        "\n",
        "# 시각화에 필요한 라이브러리 불러오기 \n",
        "import matplotlib as mpl\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns \n",
        "import matplotlib.ticker as ticker\n",
        "\n",
        "# 분류모델 학습에 필요한 라이브러리 불러오기 \n",
        "from sklearn.svm import SVC \n",
        "from sklearn.model_selection import train_test_split \n",
        "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score  \n",
        "from sklearn.model_selection import cross_val_score, KFold \n",
        "from sklearn.metrics import average_precision_score \n",
        "from sklearn.metrics import confusion_matrix \n",
        "from sklearn.metrics import precision_recall_curve, plot_precision_recall_curve  \n",
        "from sklearn.metrics import roc_auc_score, plot_roc_curve "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-2: 데이터 불러오기]"
      ],
      "metadata": {
        "id": "yG-8prZRIbMn"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aDjTcm7mQP1t",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "outputId": "e3890631-2460-40e1-e0b9-754ba5337698"
      },
      "source": [
        "df = pd.read_csv(\"https://raw.githubusercontent.com/skku-ai-textbook/aitextbook/main/data/CH13-data1.csv\")\n",
        "print(\"데이터 셋은 총 {}개의 행과 {}개의 열로 이루어져있습니다\".format(*df.shape))\n",
        "\n",
        "# 첫 5행 출력\n",
        "df.head(5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "데이터 셋은 총 5000개의 행과 10개의 열로 이루어져있습니다\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-041b10d3-59d8-48bc-95ca-6e1b28729476\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>gender</th>\n",
              "      <th>income</th>\n",
              "      <th>edu</th>\n",
              "      <th>city</th>\n",
              "      <th>smartphone</th>\n",
              "      <th>sns</th>\n",
              "      <th>movieNm</th>\n",
              "      <th>theaterNm</th>\n",
              "      <th>newspaper</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>33</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>79</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>49</td>\n",
              "      <td>0</td>\n",
              "      <td>8</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>31</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>12</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>55</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-041b10d3-59d8-48bc-95ca-6e1b28729476')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-041b10d3-59d8-48bc-95ca-6e1b28729476 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-041b10d3-59d8-48bc-95ca-6e1b28729476');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   age  gender  income  edu  ...  sns  movieNm  theaterNm  newspaper\n",
              "0   33       0       6    6  ...    0        4          1          0\n",
              "1   79       1       1    2  ...    0        0          0          0\n",
              "2   49       0       8    5  ...    0        5          0          1\n",
              "3   31       1       1    5  ...    1       12          0          0\n",
              "4   55       0       6    5  ...    1        0          0          1\n",
              "\n",
              "[5 rows x 10 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-3: 신문기사 열독자 수 확인 및 결측값 확인]"
      ],
      "metadata": {
        "id": "mt60mxMt1s5V"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bosCg7pnQP1w",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "90898c88-ddff-48c2-ba4b-424f1c33eac6"
      },
      "source": [
        "# 신문기사 열독 여부 확인 \n",
        "print(df['newspaper'].value_counts())\n",
        "\n",
        "# 결측값이 있는 행의 수 확인\n",
        "print('\\n결측값이 있는 행의 수: {}'.format(df['newspaper'].isnull().sum()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1    2667\n",
            "0    2333\n",
            "Name: newspaper, dtype: int64\n",
            "\n",
            "결측값이 있는 행의 수: 0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-4: 분류 모델 실습에 사용할 데이터 시각화] "
      ],
      "metadata": {
        "id": "bU7VR1tr4b0G"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vvbMUdSeQP1z",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 820
        },
        "outputId": "60dfe623-1795-4dc6-98a4-20621720590b"
      },
      "source": [
        "# 캔버스 설정 \n",
        "figure, axs = plt.subplots(nrows=3, ncols=3) \n",
        "figure.set_size_inches(16,16) \n",
        "\n",
        "# 신문기사 열독 여부별 나이 특성을 분포도로 표시\n",
        "sns.histplot(data=df, x='age', hue='newspaper', ax=axs[0,0]) \n",
        "\n",
        "# 각 특성별의 신문기사 열독 여부 차이 시각화 \n",
        "for i, cols in enumerate(df.columns[1:9]): # 각 특성 선택\n",
        "\n",
        "  # 8개의 이미지 배치를 위한 i값 설정 \n",
        "  i=i+1 # 그래프 배치   \n",
        "  ax = sns.countplot(data=df, x=cols, hue='newspaper', ax=axs[i//3, i%3]) \n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x1152 with 9 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-5: 학습 및 검증 데이터 준비] "
      ],
      "metadata": {
        "id": "DVoIg9Cw4vyp"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TKap-SkSQP18",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c4ddef02-bb68-4916-8e84-1020c3ebcdac"
      },
      "source": [
        "# 데이터 준비\n",
        "X = df.drop(columns=['newspaper']) # 예측에 활용한 데이터 제거 \n",
        "Y = df['newspaper'] #예측할 변인 저장\n",
        "\n",
        "# 데이터셋을 훈련과 검증 데이터 셋으로 분리 \n",
        "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.25, random_state=1)\n",
        "\n",
        "# 데이터셋 크기 출력 \n",
        "print(\"전체데이터셋의 크기: {0}\\n학습데이터셋의 크기: {1} \\n검증데이터셋의 크기: {2}\\n검증데이터셋의 비율: {3}%\"\\\n",
        "      .format(len(df), len(X_train), len(X_test), len(X_test)/len(df)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "전체데이터셋의 크기: 5000\n",
            "학습데이터셋의 크기: 3750 \n",
            "검증데이터셋의 크기: 1250\n",
            "검증데이터셋의 비율: 0.25%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-6: 분류 모델 학습] "
      ],
      "metadata": {
        "id": "5dwzBAB487Ho"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uKgQ22RiQP19",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a690e6d0-9d1d-40be-aed0-4712ffc9064e"
      },
      "source": [
        "# 학습에 사용할 특성들 지정 \n",
        "list_of_features = ['smartphone','sns'] \n",
        "\n",
        "# 모델 학습 \n",
        "print('총 {feat} 개의 특성을 사용하여 신문구독자 분류 모델을 학습합니다'\\\n",
        "      .format(feat=len(X_train[list_of_features].columns))) # 사용되는 특성 갯수 출력\n",
        "\n",
        "# 학습 모델 선언 \n",
        "svm = SVC(probability=True)\n",
        "\n",
        "# 학습 \n",
        "svm.fit(X_train[list_of_features], y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "총 2 개의 특성을 사용하여 신문구독자 분류 모델을 학습합니다\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SVC(probability=True)"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-7: 혼동 행렬 만들기] "
      ],
      "metadata": {
        "id": "WlfvXiuM91b0"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QqJz2zR8QP19",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bea12c63-7d16-4efe-86e9-33e410a7af2a"
      },
      "source": [
        "# 검증 데이터셋을 결과 예측\n",
        "preds = svm.predict(X_test[list_of_features])\n",
        "cf_matrix = confusion_matrix(y_test, preds)\n",
        "\n",
        "\n",
        "# 혼동 행렬 출력 \n",
        "print(cf_matrix)\n",
        "\n",
        "# 혼동 행렬 값을 변수에 저장\n",
        "tn, fp, fn, tp = (cf_matrix[0,0], cf_matrix[0, 1], \n",
        "                  cf_matrix[1,0], cf_matrix[1,1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[180 402]\n",
            " [  5 663]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-8: 실제 신문기사 열독자 및 비열독자 사례 수 출력]"
      ],
      "metadata": {
        "id": "TUOJQdMG_sOg"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "At_87oNzQP1-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c774793c-ddbe-46b5-e959-aa02f19e11a6"
      },
      "source": [
        "# 실제 신문기사 열독자 수 확인 \n",
        "\n",
        "print(\"검증데이터셋에 포함된 응답자 중 실제 신문기사를 읽고 있다고 대답한 응답자의 수: {}\".\\\n",
        "      format(len(y_test[y_test==1])))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "검증데이터셋에 포함된 응답자 중 실제 신문기사를 읽고 있다고 대답한 응답자의 수: 668\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-9: 혼동행렬 결과를 활용하여 분류모델 평가지표 구하기] "
      ],
      "metadata": {
        "id": "3CfwkwXnDcIl"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WIzMnbMwQP1-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4c6085f7-4b87-44c8-f217-dd8dd8595458"
      },
      "source": [
        "# 평가지표 계산 \n",
        "accuracy = (tp+tn)/(tp+fp+tn+fn) # 정확도 \n",
        "precision = (tp)/(tp+fp) # 정밀도 \n",
        "recall  = (tp)/(tp+fn) # 재현도 \n",
        "f1 = 2*(precision*recall)/(precision+recall) #F-1\n",
        "\n",
        "# 평가 지표 출력 \n",
        "print(\"모델 정확도     : {0:.3f}\".format(accuracy))\n",
        "print(\"모델 정밀도     : {0:.3f}\".format(precision))\n",
        "print(\"모델 재현도     : {0:.3f}\".format(recall))\n",
        "print(\"모델 F-1 스코어 : {0:.3f}\".format(f1))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "모델 정확도     : 0.674\n",
            "모델 정밀도     : 0.623\n",
            "모델 재현도     : 0.993\n",
            "모델 F-1 스코어 : 0.765\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7HZR8L0NQP1-"
      },
      "source": [
        "### [코드 예제 14-10: sklearn의 함수를 이용하여 분류모델 평가지표 구하기]"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t0WkwwYpQP1-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "390812b1-1c07-462f-b813-5069dd485a3f"
      },
      "source": [
        "# 평가지표 구하기 \n",
        "accuracy_svm1 = accuracy_score(y_test, preds)\n",
        "precision_svm1 = precision_score(y_test, preds)\n",
        "recall_svm1 = recall_score(y_test, preds)\n",
        "f1_svm1 = f1_score(y_test, preds)\n",
        "auc_svm1 = roc_auc_score(y_test, preds)\n",
        "\n",
        "# 평가지표 출력\n",
        "print(\"모델 정확도    : {s:8.3f}\".format(s=accuracy_svm1))\n",
        "print(\"모델 정밀도    : {s:8.3f}\".format(s=precision_svm1))\n",
        "print(\"모델 재현도    : {s:8.3f}\".format(s=recall_svm1))\n",
        "print(\"모델 F-1스코어 : {s:8.3f}\".format(s=f1_svm1))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "모델 정확도    :    0.674\n",
            "모델 정밀도    :    0.623\n",
            "모델 재현도    :    0.993\n",
            "모델 F-1스코어 :    0.765\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-11: 임계 값에 따라 신문기사 열독자로 분류되는 응답자의 수 구하기]"
      ],
      "metadata": {
        "id": "72xlYJrwCrAX"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5tJFEHQmQP1_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "24151989-0c2e-46dc-f872-00611285d430"
      },
      "source": [
        "# 신문기사 열독자일 확률값 계산 \n",
        "probs=pd.DataFrame(svm.predict_proba(X_test[list_of_features]))[0]\n",
        "probs.head()\n",
        "\n",
        "# 임계 값에 따라 신문기사 열독자로 분류되는 응답자의 수를 계산하기 \n",
        "for i in np.arange(0,1.1,.1):\n",
        "    print(\"임계 값을 {0:.0f}%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 {1}\".\\\n",
        "          format(i*100, len(probs[probs>=i])))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "임계값을 0%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 1250\n",
            "임계값을 10%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 1250\n",
            "임계값을 20%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 1250\n",
            "임계값을 30%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 1250\n",
            "임계값을 40%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 50%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 60%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 70%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 80%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 90%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 185\n",
            "임계값을 100%로 설정 했을 때 신문기사 열독자로 분류되는 응답자의 수 0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-12: plot_roc_curve를 사용한 ROC 커브 그리기]"
      ],
      "metadata": {
        "id": "Gy20VtODE9J0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#plot_roc_curve 펑션을 통한 ROC커브 그리기 \n",
        "plot_roc_curve(svm, X_test[list_of_features], y_test) \n",
        "plt.show()       "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        },
        "id": "JcamIL47DTx4",
        "outputId": "1bcc38d1-d8d3-4aed-bcaa-7b30c2df79dd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_roc_curve is deprecated; Function :func:`plot_roc_curve` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: :meth:`sklearn.metric.RocCurveDisplay.from_predictions` or :meth:`sklearn.metric.RocCurveDisplay.from_estimator`.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### [코드 예제 14-13: 인구통계학적 모델을 추가한 분류모델]"
      ],
      "metadata": {
        "id": "BLg5NzbpGcCd"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M6eYasPUQP2A",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e2b787a7-cf2a-4ae9-fd84-0c850fadb7d0"
      },
      "source": [
        "# 특성 추가 \n",
        "list_of_features2 = ['smartphone','sns','age', 'gender','income','edu','city']\n",
        "\n",
        "# 모델 학습 \n",
        "svm2 = SVC(probability=True)\n",
        "svm2.fit(X_train[list_of_features2], y_train)\n",
        "\n",
        "# 검증 및 성능 평가 \n",
        "preds2 = svm2.predict(X_test[list_of_features2]) \n",
        "print(confusion_matrix(y_test, preds2))\n",
        "\n",
        "# 정확도, 정밀도, 재현도, F1-score등 평가지표 구하기 \n",
        "accuracy_svm2 = accuracy_score(y_test, preds2) #정확도\n",
        "precision_svm2 = precision_score(y_test, preds2) #정밀도\n",
        "recall_svm2 = recall_score(y_test, preds2) #재현도\n",
        "f1_svm2 = f1_score(y_test, preds2) #F-1\n",
        "\n",
        "\n",
        "#평가지표 출력 \n",
        "print(\"\\n정확도    : {s:8.3f}\".format(s=accuracy_svm2))\n",
        "print(\"정밀도    : {s:8.3f}\".format(s=precision_svm2))\n",
        "print(\"재현도    : {s:8.3f}\".format(s=recall_svm2))\n",
        "print(\"F-1스코어 : {s:8.3f}\".format(s=f1_svm2))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[338 244]\n",
            " [105 563]]\n",
            "\n",
            "정확도    :    0.721\n",
            "정밀도    :    0.698\n",
            "재현도    :    0.843\n",
            "F-1스코어 :    0.763\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-14: 새로운 모델의 ROC 커브 그리기] "
      ],
      "metadata": {
        "id": "cmOo255oLu9B"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#plot_roc_curve 펑션을 통한 ROC커브 그리기 \n",
        "plot_roc_curve(svm2, X_test[list_of_features2], y_test) \n",
        "plt.show()       "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        },
        "id": "QqFXzP-BG-D9",
        "outputId": "c37e9925-77a9-4587-96f6-ff3250712008"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_roc_curve is deprecated; Function :func:`plot_roc_curve` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: :meth:`sklearn.metric.RocCurveDisplay.from_predictions` or :meth:`sklearn.metric.RocCurveDisplay.from_estimator`.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-15: 모든 변수를 투입한 모델 학습] "
      ],
      "metadata": {
        "id": "VzWN_L_TMiwW"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WGmz-oYMQP2D",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 455
        },
        "outputId": "5ba8682b-96f4-4095-d281-3088b227d640"
      },
      "source": [
        "# 학습에 사용할 특성 지정\n",
        "list_of_features3 = ['smartphone','sns','age', 'gender',\n",
        "                     'income','edu','city','movieNm','theaterNm']\n",
        "\n",
        "# 분류 모델 학습\n",
        "svm3= SVC(probability=True)\n",
        "svm3.fit(X_train[list_of_features3], y_train)\n",
        "\n",
        "# 분류   \n",
        "preds3 = svm3.predict(X_test[list_of_features3])\n",
        "\n",
        "# 혼동행렬 출력 \n",
        "print(confusion_matrix(y_test, preds3))\n",
        "\n",
        "# 평가지표 구하기\n",
        "accuracy_svm3 = accuracy_score(y_test, preds3)\n",
        "precision_svm3 = precision_score(y_test, preds3)\n",
        "recall_svm3 = recall_score(y_test, preds3)\n",
        "f1_svm3 = f1_score(y_test, preds3)\n",
        "auc_svm3 = roc_auc_score(y_test, preds3)\n",
        "\n",
        "#평가지표 출력 \n",
        "print(\"\\n정확도    : {s:8.3f}\".format(s=accuracy_svm3))\n",
        "print(\"정밀도    : {s:8.3f}\".format(s=precision_svm3))\n",
        "print(\"재현도    : {s:8.3f}\".format(s=recall_svm3))\n",
        "print(\"F-1스코어 : {s:8.3f}\".format(s=f1_svm3))\n",
        "\n",
        "#ROC Curve 그리기\n",
        "plot_roc_curve(svm3, X_test[list_of_features3], y_test) \n",
        "plt.show()       "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[346 236]\n",
            " [113 555]]\n",
            "\n",
            "정확도    :    0.721\n",
            "정밀도    :    0.702\n",
            "재현도    :    0.831\n",
            "F-1스코어 :    0.761\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function plot_roc_curve is deprecated; Function :func:`plot_roc_curve` is deprecated in 1.0 and will be removed in 1.2. Use one of the class methods: :meth:`sklearn.metric.RocCurveDisplay.from_predictions` or :meth:`sklearn.metric.RocCurveDisplay.from_estimator`.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### [코드 예제 14-16: 분류 모델 K-겹 교차 검증] "
      ],
      "metadata": {
        "id": "F5QtS_wwQY1-"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5M1gkmQKQP2F",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2055479b-eafa-4fbf-9f71-76c6f182a438"
      },
      "source": [
        "# K-fold 교차 검증 설정\n",
        "folds = KFold(n_splits = 5, shuffle = True, random_state = 83731)\n",
        "\n",
        "# 모델 선언 \n",
        "svc = SVC(probability=True)\n",
        "\n",
        "# 평가지표 계산 \n",
        "accuracy_SVM = cross_val_score(svc, X, Y, scoring='accuracy', cv=folds) # 정확도 \n",
        "precision_SVM = cross_val_score(svc, X, Y, scoring='precision', cv=folds) #정밀도\n",
        "recall_SVM = cross_val_score(svc, X, Y, scoring='recall', cv=folds) #재현도\n",
        "f1_SVM = cross_val_score(svc, X, Y, scoring='f1', cv=folds) #f1-스코어\n",
        "roc_auc_SVM = cross_val_score(svc, X, Y, scoring='roc_auc', cv=folds) #AUC \n",
        "\n",
        "#평가지표 출력 \n",
        "print(\"\\n정확도 평균: {s:8.3f}\".format(s=accuracy_SVM.mean()))\n",
        "print(\"정밀도 평균: {s:8.3f}\".format(s=precision_SVM.mean()))\n",
        "print(\"재현도 평균: {s:8.3f}\".format(s=recall_SVM.mean()))\n",
        "print(\"F-1스코어 평균: {s:8.3f}\".format(s=f1_SVM.mean()))\n",
        "print(\"AUC커브 평균: {s:8.3f}\".format(s=roc_auc_SVM.mean()))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "정확도 평균:    0.733\n",
            "정밀도 평균:    0.706\n",
            "재현도 평균:    0.856\n",
            "F-1스코어 평균:    0.774\n",
            "AUC커브 평균:    0.784\n"
          ]
        }
      ]
    }
  ]
}