Project1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Project1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyMLCVK5vc+AtJ+QiAe3oZD9",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/snaiws/Project-1/blob/main/Project1.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ebndyj4xpdPt"
      },
      "source": [
        "다음 분기에 어떤 게임을 설계해야 할까\n",
        "라는 고민을 해결하기 위해, 기존의 데이터에 기반하여 인사이트를 얻고, 의사결정을 하고자 하는 것이 여러분의 목표입니다.\n",
        "\n",
        "데이터셋을 읽는 것 부터,\n",
        "이러한 게임을 만들어야 한다는 결론까지 의 과정을 colab을 사용하여 작성하고,\n",
        "해당 내용을 기반으로 설명하는 영상을 작성하는 것이\n",
        "이규호님이 이번 프로젝트 기간 동안 하셔야 하는 일입니다.\n",
        "\n",
        "이규호님의 생각이나 가정들을 설명하는 과정에서 최대한 배경지식이 없는 사람들도 이해할 수 있도록 노력하셔야 할 겁니다.\n",
        "\n",
        "*지역에 따라서 선호하는 게임 장르가 다를까 라는 질문에 대답을 하셔야합니다.\n",
        "\n",
        "*연도별 게임의 트렌드가 있을까 라는 질문에 대답을 하셔야합니다.\n",
        "\n",
        "*출고량이 높은 게임에 대한 분석 및 시각화 프로세스가 포함되어야 합니다.\n",
        "\n",
        "위 3가지 내용과 여러분들의 개인적인 분석 목표, 질문을 더하여 프로젝트를 완성하세요.\n",
        "\n",
        "배운내용\n",
        "Sprint 1 : EDA, Feature Engineering, Data Manipulation, Data Visualization\n",
        "Sprint 2 : Hypothesis Test, CI, Bayesian\n",
        "Sprint 3 : DR, PCA, Cluster\n",
        "\n",
        "발표영상의 시간은 5분으로 제한합니다.\n",
        "발표영상 촬영 방법은 다른 문서를 참조하세요.\n",
        "가능한 코드에 커멘트를 달아주는 것이 권장되지만, 모든 코드 내용을 한줄 한줄 설명할 필요는 없습니다.\n",
        "단, 내가 이런 분석을 했는데 이 부분은 꼭 설명하고 싶다, 라는 내용이 있으면 설명해도 좋습니다.\n",
        "\n",
        "여러분이 생각하는 문제 (혹은 풀었으면 좋겠는 문제) 제시\n",
        "그 문제에 대한 가설 및 사용한 데이터셋에 대한 설명\n",
        "작업 결과 및 결론\n",
        "\n",
        "발표 슬라이드\n",
        "여러분이 진행하신 프로젝트를 설명하는 슬라이드를 만듭니다. 이 때 슬라이드에 들어가야 할 내용은 다음과 같습니다.\n",
        "\n",
        "여러분이 생각하는 문제 (혹은 풀었으면 좋겠는 문제) 제시\n",
        "그 문제에 대한 가설 및 사용한 데이터셋에 대한 설명\n",
        "작업 결과 및 결론\n",
        "\n",
        "어떤 모델을 어떤 코드로 짰냐 보다는,\n",
        "문제 제시 및 그 문제를 해결하기 위해 사용한 방법과 결과들을 설명하는 방식으로 이루어집니다.\n",
        "발표 시간이 5분이 넘으면 아무리 좋은 분석을 했어도 1점입니다."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 431
        },
        "id": "Mkvd4EKSWb0s",
        "outputId": "16cac068-a346-493a-da33-65aa45ce51b2"
      },
      "source": [
        "#데이터 전처리\n",
        "\n",
        "#데이터 조사 결과 Year에 천,백의자리가 빠진 성분이 있었음\n",
        "  #22이상은 1900을 더하고 22 아래는 2000을 더함\n",
        "#Sales의 숫자값 뒤에 K나 M이 붙은 값들이 있었음(그래서 원본과 대조해보니 의미없는 값)\n",
        "  #K,M제거함\n",
        "#Sales의 값이 이상하게 높아서 봤더니 원래보다 *1000된 부분이 좀 있었음 0.02 0.03 0.04 0.05와 그 이상의 몇몇 값에 1000곱한듯함\n",
        "  #세계에서 가장 많이 팔린 게임 테트리스는 없고 두 번째인 위 스포츠 Sales값을 참고하여(최대 41.49)\n",
        "  #45를 기준으로 이것보다 큰 값을 가진 수를 1000으로 나누고\n",
        "  #정확히 20, 30, 40인 값을 1000으로 나눔\n",
        "#Feature에 Sales들의 sum을 추가\n",
        "#Year, Genre, Publisher에 nan값이 있었음\n",
        "  #드랍하기엔 Sales가 100만장 이상인 항목 중 nan값을 가진게 30개가 넘음\n",
        "  #하지만 Sum_total의 0.5% 이상 차지하는 항목은 없음\n",
        "  #그래서 year를 1979로, Genre와 Publisher는 Misc로 넣기로 함.(장르에 이미 Misc가 있음)\n",
        "  #데이터 분석할 때 1979년 혹은 Misc에 해당하는 값이 유의미하게 커지면 데이터 조사를 다시 해보는걸로\n",
        "  #100만장 이상은 찾아서 기입하고 그 아래의 nan값은 Misc로 하기로 함\n",
        "\n",
        "#플랫폼이 PC, console, portable로 구분됨. Ptype 이라는 feature 추가\n",
        "\n",
        "#라이브러리 불러오기\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from scipy.stats import chisquare\n",
        "from scipy.stats import f_oneway\n",
        "from scipy.stats import chi2_contingency\n",
        "import warnings \n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "\n",
        "\n",
        "#데이터 불러오기\n",
        "url='https://raw.githubusercontent.com/snaiws/ds-section1-sprint3/master/vgames2.csv?token=ANALH6V5SVWCK6ERF4YRXI3AXFXEC'\n",
        "df=pd.read_csv(url)\n",
        "\n",
        "#Year에서 천,백의 자리 빠진 값 보충\n",
        "for i in range(len(df['Year'])):\n",
        "  if np.isnan(df['Year'][i]) == True:\n",
        "    continue\n",
        "  if df['Year'][i]<22:\n",
        "    df['Year'][i]=df['Year'][i]+2000\n",
        "  if df['Year'][i]<100:\n",
        "    df['Year'][i]=df['Year'][i]+1900\n",
        "\n",
        "#Sales의 K,M 제거 후 형변환\n",
        "def dels(z):\n",
        "  for i in range(len(z)):\n",
        "    if z[i].find('K')!=-1:\n",
        "      z[i]=float(z[i][0:z[i].find('K')])\n",
        "    elif z[i].find('M')!=-1:\n",
        "      z[i]=float(z[i][0:z[i].find('M')])\n",
        "    else:\n",
        "      z[i]=float(z[i])\n",
        "dels(df['NA_Sales'])\n",
        "dels(df['EU_Sales'])\n",
        "dels(df['JP_Sales'])\n",
        "dels(df['Other_Sales'])\n",
        "\n",
        "#Sales의 이상한 수치 수정\n",
        "def val(z):\n",
        "  for i in range(len(z)):\n",
        "    if z[i] > 45 or z[i]==40 or z[i]==30 or z[i]==20 : #이거 기준을 아웃라이어로 해야 할듯\n",
        "      z[i]=z[i]/1000\n",
        "val(df['NA_Sales'])\n",
        "val(df['EU_Sales'])\n",
        "val(df['JP_Sales'])\n",
        "val(df['Other_Sales'])\n",
        "\n",
        "#새로운 feature인 sales total 추가 후 sales total 기준 내림차순으로 정렬 후 인덱스 수정\n",
        "df['Sales_sum']=df['NA_Sales']+df['EU_Sales']+df['JP_Sales']+df['Other_Sales']\n",
        "df = df.sort_values(by='Sales_sum' ,ascending=False)\n",
        "df['index']=range(len(df['Name']))\n",
        "df=df.set_index(df['index']).loc[:,'Name':'Sales_sum']\n",
        "\n",
        "#NaN 처리\n",
        "df['Year'].fillna(1979,inplace=True)\n",
        "df['Genre'].fillna('Misc',inplace=True)\n",
        "df['Publisher'].fillna('Misc',inplace=True)\n",
        "\n",
        "#Ptype feature 추가\n",
        "df['Ptype']=df['Platform']\n",
        "df['Ptype']=df['Ptype'].replace('WS','portable').replace('GG','portable').replace('GB','portable').replace('GBA','portable').replace('DS','portable').replace('3DS','portable').replace('PSP','portable').replace('PSV','portable')\n",
        "for i in range(len(df['Ptype'])):\n",
        "  if df['Ptype'][i] != 'PC' and df['Ptype'][i] != 'portable' and df['Ptype'][i] != 'Misc':\n",
        "    df['Ptype'][i] = 'console'\n",
        "\n",
        "#object to float\n",
        "df['NA_Sales']=df['NA_Sales'].astype(float)\n",
        "df['EU_Sales']=df['EU_Sales'].astype(float)\n",
        "df['JP_Sales']=df['JP_Sales'].astype(float)\n",
        "df['Other_Sales']=df['Other_Sales'].astype(float)\n",
        "df['Sales_sum']=df['Sales_sum'].astype(float)\n",
        "\n",
        "df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Name</th>\n",
              "      <th>Platform</th>\n",
              "      <th>Year</th>\n",
              "      <th>Genre</th>\n",
              "      <th>Publisher</th>\n",
              "      <th>NA_Sales</th>\n",
              "      <th>EU_Sales</th>\n",
              "      <th>JP_Sales</th>\n",
              "      <th>Other_Sales</th>\n",
              "      <th>Sales_sum</th>\n",
              "      <th>Ptype</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>index</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Wii Sports</td>\n",
              "      <td>Wii</td>\n",
              "      <td>2006.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Nintendo</td>\n",
              "      <td>41.49</td>\n",
              "      <td>29.02</td>\n",
              "      <td>3.77</td>\n",
              "      <td>8.46</td>\n",
              "      <td>82.74</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Super Mario Bros.</td>\n",
              "      <td>NES</td>\n",
              "      <td>1985.0</td>\n",
              "      <td>Platform</td>\n",
              "      <td>Nintendo</td>\n",
              "      <td>29.08</td>\n",
              "      <td>3.58</td>\n",
              "      <td>6.81</td>\n",
              "      <td>0.77</td>\n",
              "      <td>40.24</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Mario Kart Wii</td>\n",
              "      <td>Wii</td>\n",
              "      <td>2008.0</td>\n",
              "      <td>Racing</td>\n",
              "      <td>Nintendo</td>\n",
              "      <td>15.85</td>\n",
              "      <td>12.88</td>\n",
              "      <td>3.79</td>\n",
              "      <td>3.31</td>\n",
              "      <td>35.83</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Wii Sports Resort</td>\n",
              "      <td>Wii</td>\n",
              "      <td>2009.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Nintendo</td>\n",
              "      <td>15.75</td>\n",
              "      <td>11.01</td>\n",
              "      <td>3.28</td>\n",
              "      <td>2.96</td>\n",
              "      <td>33.00</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Pokemon Red/Pokemon Blue</td>\n",
              "      <td>GB</td>\n",
              "      <td>1996.0</td>\n",
              "      <td>Role-Playing</td>\n",
              "      <td>Nintendo</td>\n",
              "      <td>11.27</td>\n",
              "      <td>8.89</td>\n",
              "      <td>10.22</td>\n",
              "      <td>1.00</td>\n",
              "      <td>31.38</td>\n",
              "      <td>portable</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16593</th>\n",
              "      <td>Mighty No. 9</td>\n",
              "      <td>XOne</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Platform</td>\n",
              "      <td>Deep Silver</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16594</th>\n",
              "      <td>XI Coliseum</td>\n",
              "      <td>PSP</td>\n",
              "      <td>2006.0</td>\n",
              "      <td>Puzzle</td>\n",
              "      <td>Sony Computer Entertainment</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>portable</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16595</th>\n",
              "      <td>Davis Cup Tennis</td>\n",
              "      <td>GBA</td>\n",
              "      <td>2002.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Ubisoft</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>portable</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16596</th>\n",
              "      <td>Guild Wars: Factions</td>\n",
              "      <td>PC</td>\n",
              "      <td>2006.0</td>\n",
              "      <td>Role-Playing</td>\n",
              "      <td>NCSoft</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.01</td>\n",
              "      <td>PC</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16597</th>\n",
              "      <td>SCORE International Baja 1000: The Official Game</td>\n",
              "      <td>PS2</td>\n",
              "      <td>2008.0</td>\n",
              "      <td>Racing</td>\n",
              "      <td>Activision</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>16598 rows × 11 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   Name  ...     Ptype\n",
              "index                                                    ...          \n",
              "0                                            Wii Sports  ...   console\n",
              "1                                     Super Mario Bros.  ...   console\n",
              "2                                        Mario Kart Wii  ...   console\n",
              "3                                     Wii Sports Resort  ...   console\n",
              "4                              Pokemon Red/Pokemon Blue  ...  portable\n",
              "...                                                 ...  ...       ...\n",
              "16593                                      Mighty No. 9  ...   console\n",
              "16594                                       XI Coliseum  ...  portable\n",
              "16595                                  Davis Cup Tennis  ...  portable\n",
              "16596                              Guild Wars: Factions  ...        PC\n",
              "16597  SCORE International Baja 1000: The Official Game  ...   console\n",
              "\n",
              "[16598 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6V7G2o6VJxQk"
      },
      "source": [
        "#함수모음\n",
        "\n",
        "def ele(z): #열 성분 중복없이 추출, input/output = 리스트\n",
        "  a=list()\n",
        "  for i in range(len(z)):\n",
        "    b=0\n",
        "    for j in range(len(a)):\n",
        "      if z[i] == a[j]:\n",
        "        b=b+1\n",
        "    if b==0:\n",
        "      a.append(z[i])\n",
        "  return a\n",
        "  \n",
        "def ct(z): #열의 모든 성분의 타입 추출\n",
        "  a=list()\n",
        "  for i in range(len(z)):\n",
        "    a.append(type(z[i]))\n",
        "  return a\n",
        "\n",
        "def standard(z): #정규화\n",
        "  z=[(x-np.mean(z))/np.std(z,ddof=1) for x in z]\n",
        "  return z\n",
        "\n",
        "def flbc(m):#find location by corr, m is df\n",
        "  a=list()\n",
        "  coldex=m.columns\n",
        "  for i in range(m.shape[0]): #세로길이\n",
        "    for j in range(m.shape[1]): #가로길이\n",
        "      if (m.iloc[i,j]>0.8 or m.iloc[i,j]<-0.8) and i>j :\n",
        "        a.append([coldex[i],coldex[j]])\n",
        "  return a\n",
        "\n",
        "def indsum(df):\n",
        "  a=ele(df.iloc[:,0])\n",
        "  c=pd.DataFrame(a)\n",
        "  for i in range(df.shape[1]-1):\n",
        "    b=list()\n",
        "    for j in range(len(a)):\n",
        "      b.append(np.sum(df.loc[df.iloc[:,0] == a[j]].iloc[:,i+1]))\n",
        "    c=pd.concat([c,pd.DataFrame(b)],axis=1)\n",
        "  return c\n",
        "\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xshT2czDRDxw"
      },
      "source": [
        "#가설검정\n",
        "#t-test : onesample : 평균이 특정값과 일치하는지 twosample : 평균이 서로 같은지\n",
        "#chi-square test ㅣ one : 분포가 특정 분포와 같은지, two : 서로 분포가 비슷한지 or 독립인지\n",
        "#ABtest\n",
        "#Cov, Corr, PCA, Clustering\n",
        "#플랏 : 라인 바 박스 스캐터\n",
        "\n",
        "#다음 분기에 어떤 게임을 설계해야 할까\n",
        "#*지역에 따라서 선호하는 게임 장르가 다를까 라는 질문에 대답을 하셔야합니다. ->지역별 sales와 장르의 연관성\n",
        "  #귀무가설 : 지역별 선호하는 게임 장르가 같다.\n",
        "  #f test\n",
        "  #장르 지역sales\n",
        "#*연도별 게임의 트렌드가 있을까 라는 질문에 대답을 하셔야합니다. 연도 vs 장르 vs 세일\n",
        "  #귀무가설 : 연도별 게임(카테고리)의 트렌드(출고량차이?변화?)는 없고 비슷비슷하다\n",
        "  #연도 장르-totalsales\n",
        "#*출고량이 높은 게임에 대한 분석 및 시각화 프로세스가 포함되어야 합니다.\n",
        "  #tsales top 100\n",
        "#어떤 장르 어떤 플랫폼 어떤 지역\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HE3Wm5Xno_Ti",
        "outputId": "4f2fa644-7ae7-409c-e255-5f584d3fc8d6"
      },
      "source": [
        "#1차 해석\n",
        "#features\n",
        "  #year(1980~2020, 1979)\n",
        "  #장르(11종)\n",
        "  #플랫폼(31종)\n",
        "  #퍼블리셔(578종)\n",
        "  #feature=['Name','Platform','Year','Genre','Publisher','NA_Sales','EU_Sales','JP_Sales','Other_Sales','Sales_sum','Ptype'];\n",
        "print(ele(df['Platform']))\n",
        "print(ele(df['Year']))\n",
        "print(ele(df['Genre']))\n",
        "print(ele(df['Ptype']))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['Wii', 'NES', 'GB', 'DS', 'X360', 'PS3', 'PS2', 'SNES', 'GBA', '3DS', 'PS4', 'N64', 'PS', 'PSP', 'PC', 'GC', 'XB', '2600', 'XOne', 'WiiU', 'GEN', 'DC', 'PSV', 'SAT', 'SCD', 'WS', 'NG', 'TG16', '3DO', 'GG', 'PCFX']\n",
            "[2006.0, 1985.0, 2008.0, 2009.0, 1996.0, 1989.0, 1984.0, 2005.0, 1999.0, 2007.0, 2010.0, 2013.0, 2004.0, 1990.0, 1988.0, 2002.0, 2001.0, 2011.0, 1998.0, 2015.0, 2012.0, 2014.0, 1992.0, 1997.0, 1993.0, 1994.0, 1982.0, 2003.0, 1986.0, 2000.0, 1979.0, 1995.0, 2016.0, 1991.0, 1981.0, 1987.0, 1980.0, 1983.0, 2020.0, 2017.0]\n",
            "['Sports', 'Platform', 'Racing', 'Role-Playing', 'Puzzle', 'Misc', 'Shooter', 'Simulation', 'Action', 'Fighting', 'Adventure', 'Strategy']\n",
            "['console', 'portable', 'PC']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "adiNjJ44N1Mx",
        "outputId": "fe5bb5b3-47b3-44be-c939-450a50dd49e6"
      },
      "source": [
        "#2차 해석\n",
        "\n",
        "#변수\n",
        "  #Platform : categorical nomial unordered qualitative variable\n",
        "  #Platform 하위 : uncontinuous quantative variable\n",
        "  #Genre : categorical nomial unordered qualitative variable\n",
        "  #Genre 하위 : uncontinuous quantative variable\n",
        "  #Publisher : categorical nomial unordered qualitative variable\n",
        "  #Publisher 하위 : uncontinuous quantative variable\n",
        "  #Year : quantitive uncontinuous variable\n",
        "  #Sales : quantitive continuous variable\n",
        "  #Ptype : categorical nomial unordered qualitative variable\n",
        "\n",
        "#비교,분석할 numerical value는 Year, tSales, rSales, number of elements of a category, bin of Sales\n",
        "\n",
        "#tSales vs category 시각화\n",
        "#tSales vs 연도 - 년도의 흐름에 따라 tSales의 분포가 증가할 것이다\n",
        "#tSales vs 장르 - 장르에 따른 tsales의 분포는 uniformly distributed 할 것이다\n",
        "#tSales vs 플랫폼 - 플랫폼에 따른 tsales의 분포는 uniformly distributed 할 것이다\n",
        "#tSales vs 퍼블리셔 - 퍼블리셔에 따른 tsales의 분포는 uniformly distributed 할 것이다\n",
        "\n",
        "#totalsale-categoryA vs categoryB -> f test \n",
        "#귀무가설 : A별 게임B의 트렌드가 없고 일정하다\n",
        "\n",
        "#t플랫폼 vs 장르 - 기각\n",
        "df2= df.loc[:,['Genre','Platform','Sales_sum']]\n",
        "df2 = df2.pivot_table(index = 'Genre', columns = 'Platform', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df2.dropna(axis=1,inplace=True)\n",
        "tPlvsG=df2.corr(method='pearson', min_periods=1) #성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 플랫폼은 장르에 따른 매출이 비슷하다 라는 것\n",
        "tPlvsGcorr=flbc(tPlvsG) # 강한 상관관계가 있는 것들\n",
        "print('tPlvsGcorr=',tPlvsGcorr)\n",
        "tPlvsG_F=f_oneway(df2['3DS'],df2['DS'],df2['GBA'],df2['GC'],df2['N64'],df2['PC'],df2['PS'],df2['PS2'],df2['PS3'],df2['PS4'],df2['PSP'],df2['PSV'],df2['SAT'],df2['SNES'],df2['Wii'],df2['WiiU'],df2['X360'],df2['XB'])\n",
        "\n",
        "#tPtype vs 장르 - 기각\n",
        "df5= df.loc[:,['Genre','Ptype','Sales_sum']]\n",
        "df5 = df5.pivot_table(index = 'Genre', columns = 'Ptype', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df5.dropna(axis=1,inplace=True)\n",
        "tPtvsG=df5.corr(method='pearson', min_periods=1)\n",
        "tPtvsGcorr=flbc(tPtvsG) # 강한 상관관계가 없다\n",
        "print('tPtvsGcorr=',tPtvsGcorr)\n",
        "tPtvsG_F=f_oneway(df5['PC'],df5['console'],df5['portable'])\n",
        "\n",
        "#tPtype vs 년도 - 기각\n",
        "df7= df.loc[:,['Year','Ptype','Sales_sum']]\n",
        "df7 = df7.pivot_table(index = 'Year', columns = 'Ptype', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df7.dropna(axis=0,inplace=True)\n",
        "tPtvsY=df7.corr(method='pearson', min_periods=1)\n",
        "tPtvsYcorr=flbc(tPtvsY) # 강한 상관관계가 없다\n",
        "print('tPtvsYcorr=',tPtvsYcorr)\n",
        "tPtvsY_F=f_oneway(df7['PC'],df7['console'],df7['portable'])\n",
        "\n",
        "#t장르 vs 년도 - 기각x\n",
        "df10= df.loc[:,['Year','Genre','Sales_sum']]\n",
        "df10 = df10.pivot_table(index = 'Year', columns = 'Genre', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df10.dropna(axis=0,inplace=True)\n",
        "tGvsY=df10.corr(method='pearson', min_periods=1)\n",
        "tGvsYcorr=flbc(tGvsY) # 강한 상관관계가 없다\n",
        "print('tGvsYcorr=',tGvsYcorr)\n",
        "tGvsY_F=f_oneway(df10['Action'],df10['Adventure'],df10['Fighting'],df10['Platform'],df10['Puzzle'],df10['Racing'],df10['Role-Playing'],df10['Shooter'],df10['Simulation'],df10['Sports'],df10['Strategy'])\n",
        "\n",
        "#t장르 vs Ptype\n",
        "df11= df.loc[:,['Ptype','Genre','Sales_sum']]\n",
        "df11 = df11.pivot_table(index = 'Ptype', columns = 'Genre', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df11.dropna(axis=1,inplace=True) #드랍 필요없음\n",
        "tGvsPt=df11.corr(method='pearson', min_periods=1)#성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 장르는 Ptype에 따른 매출이 비슷하다 라는 것\n",
        "tGvsPtcorr=flbc(tGvsPt) # 강한 상관관계가 있다\n",
        "print('tGvsPtcorr=',tGvsPtcorr)\n",
        "tGvsPt_F=f_oneway(df11['Action'],df11['Adventure'],df11['Fighting'],df11['Platform'],df11['Puzzle'],df11['Racing'],df11['Role-Playing'],df11['Shooter'],df11['Simulation'],df11['Sports'],df11['Strategy'])\n",
        "\n",
        "'''\n",
        "#t플랫폼 vs 퍼블리셔 - 퍼블리셔에 따른 플랫폼판매량의 분포는 비슷하다 -> 기각시 비슷하지 않다 의미있는 통계다\n",
        "df3= df.loc[:,['Publisher','Platform','Sales_sum']]\n",
        "df3 = df3.pivot_table(index = 'Publisher', columns = 'Platform', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df3.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tPlvsPu=df3.corr(method='pearson', min_periods=1) #성분간 강한 연관이 있는게 없다\n",
        "tPlvsPucorr=flbc(tPlvsPu) # 강한 상관관계가 없다\n",
        "print('tPlvsPucorr=',tPlvsPucorr)\n",
        "#t플랫폼 vs 년도 - 년도에 따른 플랫폼판매량의 분포는 비슷하다 -> 기각시 비슷하지 않다 의미있는 통계다\n",
        "df4= df.loc[:,['Year','Platform','Sales_sum']]\n",
        "df4 = df4.pivot_table(index = 'Year', columns = 'Platform', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df4.dropna(axis=0,inplace=True)#너무 많은 NaN값 존재\n",
        "tPlvsY=df4.corr(method='pearson', min_periods=1)\n",
        "tPlvsYcorr=flbc(tPlvsY) # 강한 상관관계가 없다\n",
        "print('tPlvsYcorr=',tPlvsYcorr)\n",
        "#tPtype vs 퍼블리셔\n",
        "df6= df.loc[:,['Publisher','Ptype','Sales_sum']]\n",
        "df6 = df6.pivot_table(index = 'Publisher', columns = 'Ptype', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df6.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tPtvsPu=df6.corr(method='pearson', min_periods=1)\n",
        "tPtvsPucorr=flbc(tPtvsPu) # 강한 상관관계가 없다\n",
        "print('tPtvsPucorr=',tPtvsPucorr)\n",
        "#t장르 vs 플랫폼 - \n",
        "df8= df.loc[:,['Platform','Genre','Sales_sum']]\n",
        "df8 = df8.pivot_table(index = 'Platform', columns = 'Genre', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df8.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tGvsPl=df8.corr(method='pearson', min_periods=1)\n",
        "tGvsPlcorr=flbc(tGvsPl) # 강한 상관관계가 없다\n",
        "print('tGvsPlcorr=',tGvsPlcorr)\n",
        "#t장르 vs 퍼블리셔\n",
        "df9= df.loc[:,['Publisher','Genre','Sales_sum']]\n",
        "df9 = df9.pivot_table(index = 'Publisher', columns = 'Genre', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df9.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tGvsPu=df9.corr(method='pearson', min_periods=1)\n",
        "tGvsPucorr=flbc(tGvsPu) # 강한 상관관계가 없다\n",
        "print('tGvsPucorr=',tGvsPucorr)\n",
        "#t퍼블리셔 vs 플랫폼\n",
        "df12= df.loc[:,['Platform','Publisher','Sales_sum']]\n",
        "df12 = df12.pivot_table(index = 'Platform', columns = 'Publisher', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df12.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tPuvsPl=df12.corr(method='pearson', min_periods=1)\n",
        "tPuvsPlcorr=flbc(tPuvsPl) # 강한 상관관계가 없다\n",
        "print('tPuvsPlcorr=',tPuvsPlcorr)\n",
        "#t퍼블리셔 vs 장르\n",
        "df13= df.loc[:,['Genre','Publisher','Sales_sum']]\n",
        "df13 = df13.pivot_table(index = 'Genre', columns = 'Publisher', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df13.dropna(axis=1,inplace=True)\n",
        "tPuvsG=df13.corr(method='pearson', min_periods=1)#성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 퍼블리셔는 장르에 따른 매출이 비슷하다 라는 것\n",
        "tPuvsGcorr=flbc(tPuvsG) # 강한 상관관계가 있다\n",
        "print('tPuvsGcorr=',tPuvsGcorr)\n",
        "#t퍼블리셔 vs 년도\n",
        "df14= df.loc[:,['Year','Publisher','Sales_sum']]\n",
        "df14 = df14.pivot_table(index = 'Year', columns = 'Publisher', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df14.dropna(axis=0,inplace=True)#너무 많은 NaN값 존재\n",
        "tPuvsY=df14.corr(method='pearson', min_periods=1)\n",
        "tPuvsYcorr=flbc(tPuvsY) # 강한 상관관계가 없다\n",
        "print('tPuvsYcorr=',tPuvsYcorr)\n",
        "#t퍼블리셔 vs Ptype\n",
        "df15= df.loc[:,['Ptype','Publisher','Sales_sum']]\n",
        "df15 = df15.pivot_table(index = 'Ptype', columns = 'Publisher', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df15.dropna(axis=1,inplace=True)\n",
        "tPuvsPt=df15.corr(method='pearson', min_periods=1)#성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 퍼블리셔는 Ptype에 따른 매출이 비슷하다 라는 것\n",
        "tPuvsPtcorr=flbc(tPuvsPt) # 강한 상관관계가 있다\n",
        "print('tPuvsPtcorr=',tPuvsPtcorr)\n",
        "\n",
        "\n",
        "#t년도 vs 플랫폼 ★ NaN값이 매우 많다, 플랫폼의 수명 을 보여준다\n",
        "df16= df.loc[:,['Platform','Year','Sales_sum']]\n",
        "df16 = df16.pivot_table(index = 'Platform', columns = 'Year', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df16.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tYvsPl=df16.corr(method='pearson', min_periods=1)\n",
        "tYvsPlcorr=flbc(tYvsPl) # 강한 상관관계가 없다\n",
        "print('tYvsPlcorr=',tYvsPlcorr)\n",
        "#t년도 vs 퍼블리셔★\n",
        "df18= df.loc[:,['Publisher','Year','Sales_sum']]\n",
        "df18 = df18.pivot_table(index = 'Publisher', columns = 'Year', values = 'Sales_sum',aggfunc=np.sum)\n",
        "#df18.dropna(axis=1,inplace=True)#너무 많은 NaN값 존재\n",
        "tYvsPu=df18.corr(method='pearson', min_periods=1)\n",
        "tYvsPucorr=flbc(tYvsPu) # 강한 상관관계가 없다\n",
        "print('tYvsPucorr=',tYvsPucorr)\n",
        "'''\n",
        "\n",
        "#t년도 vs 장르 ★\n",
        "df17= df.loc[:,['Genre','Year','Sales_sum']]\n",
        "df17 = df17.pivot_table(index = 'Genre', columns = 'Year', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df17.dropna(axis=1,inplace=True)\n",
        "tYvsG=df17.corr(method='pearson', min_periods=1)#성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 년도는 장르에 따른 매출이 비슷하다 라는 것\n",
        "tYvsGcorr=flbc(tYvsG) # 강한 상관관계가 있다\n",
        "print('tYvsGcorr=',tYvsGcorr)\n",
        "tYvsG_F=f_oneway(df17[1991],df17[1992],df17[1993],df17[1994],df17[1995],df17[1996],df17[1997],df17[1998],df17[1999],df17[2000],df17[2001],df17[2002],df17[2003],df17[2004],df17[2005],df17[2006],df17[2007],df17[2008],df17[2009],df17[2010],df17[2011],df17[2012],df17[2013],df17[2014],df17[2015])\n",
        "\n",
        "#t년도 vs Ptype★\n",
        "df19= df.loc[:,['Ptype','Year','Sales_sum']]\n",
        "df19 = df19.pivot_table(index = 'Ptype', columns = 'Year', values = 'Sales_sum',aggfunc=np.sum)\n",
        "df19.dropna(axis=1,inplace=True)\n",
        "tYvsPt=df19.corr(method='pearson', min_periods=1)#성분간 강한 연관이 있는게 있다. 이것의 의미는 그 두 년도는 Ptype에 따른 매출이 비슷하다 라는 것\n",
        "tYvsPtcorr=flbc(tYvsPt) # 강한 상관관계가 있다\n",
        "print('tYvsPtcorr=',tYvsPtcorr)\n",
        "tYvsPt_F=f_oneway(df19[1985],df19[1988],df19[1992],df19[1994],df19[1995],df19[1996],df19[1997],df19[1998],df19[1999],df19[2000],df19[2001],df19[2002],df19[2003],df19[2004],df19[2005],df19[2006],df19[2007],df19[2008],df19[2009],df19[2010],df19[2011],df19[2012],df19[2013],df19[2014],df19[2015],df19[2016])\n",
        "\n",
        "\n",
        "#regionsale vs category - f test\n",
        "#귀무가설 : 지역에 따라 선호 게임(category)가 같다\n",
        "\n",
        "#rs vs 플랫폼 - 기각\n",
        "df20 = df.loc[:,['Platform','NA_Sales','EU_Sales','JP_Sales','Other_Sales']]\n",
        "df20=indsum(df20)\n",
        "df20.columns=['Platform','NA_Sales','EU_Sales','JP_Sales','Other_Sales']\n",
        "df20.set_index('Platform',inplace=True)\n",
        "PlvsRS_F=f_oneway(df20['NA_Sales'],df20['EU_Sales'],df20['JP_Sales'],df20['Other_Sales'])\n",
        "#rs vs 장르 ★ 기각\n",
        "df21 = df.loc[:,['Genre','NA_Sales','EU_Sales','JP_Sales','Other_Sales']]\n",
        "df21=indsum(df21)\n",
        "df21.columns=['Genre','NA_Sales','EU_Sales','JP_Sales','Other_Sales']\n",
        "df21.set_index('Genre',inplace=True)\n",
        "GvsRS_F=f_oneway(df21['NA_Sales'],df21['EU_Sales'],df21['JP_Sales'],df21['Other_Sales'])\n",
        "#rs vs 년도 - 기각\n",
        "df22 = df.loc[:,['Year','NA_Sales','EU_Sales','JP_Sales','Other_Sales']]\n",
        "df22=indsum(df22)\n",
        "df22.columns=['Year','NA_Sales','EU_Sales','JP_Sales','Other_Sales']\n",
        "df22.set_index('Year',inplace=True)\n",
        "YvsRS_F=f_oneway(df22['NA_Sales'],df22['EU_Sales'],df22['JP_Sales'],df22['Other_Sales'])\n",
        "#rs vs 퍼블리셔 - 기각\n",
        "df23 = df.loc[:,['Publisher','NA_Sales','EU_Sales','JP_Sales','Other_Sales']]\n",
        "df23=indsum(df23)\n",
        "df23.columns=['Publisher','NA_Sales','EU_Sales','JP_Sales','Other_Sales']\n",
        "df23.set_index('Publisher',inplace=True)\n",
        "PuvsRS_F=f_oneway(df23['NA_Sales'],df23['EU_Sales'],df23['JP_Sales'],df23['Other_Sales'])\n",
        "#rs vs Ptype - 기각x\n",
        "df24 = df.loc[:,['Ptype','NA_Sales','EU_Sales','JP_Sales','Other_Sales']]\n",
        "df24=indsum(df24)\n",
        "df24.columns=['Ptype','NA_Sales','EU_Sales','JP_Sales','Other_Sales']\n",
        "df24.set_index('Ptype',inplace=True)\n",
        "PtvsRS_F=f_oneway(df24['NA_Sales'],df24['EU_Sales'],df24['JP_Sales'],df24['Other_Sales'])\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tPlvsGcorr= [['GBA', '3DS'], ['PS2', 'GC'], ['PS2', 'PS'], ['PS3', 'GC'], ['PS4', 'PS3'], ['PSP', 'PS'], ['PSP', 'PS2'], ['PSV', 'PSP'], ['WiiU', 'GC'], ['X360', 'PS3'], ['X360', 'PS4'], ['XB', 'PS3'], ['XB', 'PS4'], ['XB', 'X360']]\n",
            "tPtvsGcorr= []\n",
            "tPtvsYcorr= [['portable', 'console']]\n",
            "tGvsYcorr= [['Misc', 'Adventure'], ['Shooter', 'Action'], ['Simulation', 'Misc'], ['Simulation', 'Puzzle'], ['Sports', 'Misc']]\n",
            "tGvsPtcorr= [['Fighting', 'Action'], ['Misc', 'Action'], ['Misc', 'Adventure'], ['Misc', 'Fighting'], ['Platform', 'Action'], ['Platform', 'Adventure'], ['Platform', 'Fighting'], ['Platform', 'Misc'], ['Puzzle', 'Adventure'], ['Racing', 'Action'], ['Racing', 'Fighting'], ['Racing', 'Misc'], ['Racing', 'Platform'], ['Role-Playing', 'Adventure'], ['Role-Playing', 'Misc'], ['Role-Playing', 'Platform'], ['Role-Playing', 'Puzzle'], ['Shooter', 'Action'], ['Shooter', 'Fighting'], ['Shooter', 'Misc'], ['Shooter', 'Platform'], ['Shooter', 'Racing'], ['Simulation', 'Adventure'], ['Simulation', 'Platform'], ['Simulation', 'Puzzle'], ['Simulation', 'Role-Playing'], ['Sports', 'Action'], ['Sports', 'Fighting'], ['Sports', 'Misc'], ['Sports', 'Platform'], ['Sports', 'Racing'], ['Sports', 'Shooter'], ['Strategy', 'Action'], ['Strategy', 'Fighting'], ['Strategy', 'Racing'], ['Strategy', 'Shooter'], ['Strategy', 'Sports']]\n",
            "tYvsGcorr= [[1994.0, 1993.0], [1995.0, 1993.0], [1999.0, 1996.0], [2000.0, 1998.0], [2001.0, 1997.0], [2002.0, 1997.0], [2002.0, 2001.0], [2003.0, 1997.0], [2003.0, 1998.0], [2003.0, 2000.0], [2003.0, 2001.0], [2003.0, 2002.0], [2004.0, 1997.0], [2004.0, 2002.0], [2004.0, 2003.0], [2006.0, 2000.0], [2007.0, 1979.0], [2007.0, 2005.0], [2007.0, 2006.0], [2008.0, 1979.0], [2008.0, 2003.0], [2008.0, 2004.0], [2008.0, 2005.0], [2008.0, 2007.0], [2009.0, 1979.0], [2009.0, 2002.0], [2009.0, 2005.0], [2009.0, 2006.0], [2009.0, 2007.0], [2009.0, 2008.0], [2010.0, 1979.0], [2010.0, 2004.0], [2010.0, 2006.0], [2010.0, 2007.0], [2010.0, 2008.0], [2010.0, 2009.0], [2011.0, 1979.0], [2011.0, 2002.0], [2011.0, 2004.0], [2011.0, 2007.0], [2011.0, 2008.0], [2011.0, 2009.0], [2011.0, 2010.0], [2012.0, 2002.0], [2012.0, 2004.0], [2012.0, 2011.0], [2013.0, 1979.0], [2013.0, 2002.0], [2013.0, 2004.0], [2013.0, 2008.0], [2013.0, 2009.0], [2013.0, 2010.0], [2013.0, 2011.0], [2013.0, 2012.0], [2014.0, 1979.0], [2014.0, 2002.0], [2014.0, 2004.0], [2014.0, 2008.0], [2014.0, 2009.0], [2014.0, 2010.0], [2014.0, 2011.0], [2014.0, 2012.0], [2014.0, 2013.0], [2015.0, 2002.0], [2015.0, 2004.0], [2015.0, 2010.0], [2015.0, 2011.0], [2015.0, 2012.0], [2015.0, 2013.0], [2015.0, 2014.0]]\n",
            "tYvsPtcorr= [[1985.0, 1979.0], [1988.0, 1979.0], [1988.0, 1985.0], [1992.0, 1979.0], [1992.0, 1985.0], [1992.0, 1988.0], [1994.0, 1979.0], [1994.0, 1985.0], [1994.0, 1988.0], [1994.0, 1992.0], [1995.0, 1979.0], [1995.0, 1985.0], [1995.0, 1988.0], [1995.0, 1992.0], [1995.0, 1994.0], [1996.0, 1979.0], [1996.0, 1985.0], [1996.0, 1988.0], [1996.0, 1992.0], [1996.0, 1994.0], [1996.0, 1995.0], [1997.0, 1979.0], [1997.0, 1985.0], [1997.0, 1988.0], [1997.0, 1992.0], [1997.0, 1994.0], [1997.0, 1995.0], [1997.0, 1996.0], [1998.0, 1979.0], [1998.0, 1985.0], [1998.0, 1988.0], [1998.0, 1992.0], [1998.0, 1994.0], [1998.0, 1995.0], [1998.0, 1996.0], [1998.0, 1997.0], [1999.0, 1979.0], [1999.0, 1985.0], [1999.0, 1988.0], [1999.0, 1992.0], [1999.0, 1994.0], [1999.0, 1995.0], [1999.0, 1996.0], [1999.0, 1997.0], [1999.0, 1998.0], [2000.0, 1979.0], [2000.0, 1985.0], [2000.0, 1988.0], [2000.0, 1992.0], [2000.0, 1994.0], [2000.0, 1995.0], [2000.0, 1996.0], [2000.0, 1997.0], [2000.0, 1998.0], [2000.0, 1999.0], [2001.0, 1979.0], [2001.0, 1985.0], [2001.0, 1988.0], [2001.0, 1992.0], [2001.0, 1994.0], [2001.0, 1995.0], [2001.0, 1996.0], [2001.0, 1997.0], [2001.0, 1998.0], [2001.0, 1999.0], [2001.0, 2000.0], [2002.0, 1979.0], [2002.0, 1985.0], [2002.0, 1988.0], [2002.0, 1992.0], [2002.0, 1994.0], [2002.0, 1995.0], [2002.0, 1996.0], [2002.0, 1997.0], [2002.0, 1998.0], [2002.0, 1999.0], [2002.0, 2000.0], [2002.0, 2001.0], [2003.0, 1979.0], [2003.0, 1985.0], [2003.0, 1988.0], [2003.0, 1992.0], [2003.0, 1994.0], [2003.0, 1995.0], [2003.0, 1996.0], [2003.0, 1997.0], [2003.0, 1998.0], [2003.0, 1999.0], [2003.0, 2000.0], [2003.0, 2001.0], [2003.0, 2002.0], [2004.0, 1979.0], [2004.0, 1985.0], [2004.0, 1988.0], [2004.0, 1992.0], [2004.0, 1994.0], [2004.0, 1995.0], [2004.0, 1996.0], [2004.0, 1997.0], [2004.0, 1998.0], [2004.0, 1999.0], [2004.0, 2000.0], [2004.0, 2001.0], [2004.0, 2002.0], [2004.0, 2003.0], [2005.0, 1992.0], [2005.0, 2004.0], [2006.0, 1979.0], [2006.0, 1985.0], [2006.0, 1988.0], [2006.0, 1992.0], [2006.0, 1994.0], [2006.0, 1995.0], [2006.0, 1996.0], [2006.0, 1997.0], [2006.0, 1998.0], [2006.0, 1999.0], [2006.0, 2000.0], [2006.0, 2001.0], [2006.0, 2002.0], [2006.0, 2003.0], [2006.0, 2004.0], [2006.0, 2005.0], [2007.0, 1979.0], [2007.0, 1985.0], [2007.0, 1988.0], [2007.0, 1992.0], [2007.0, 1994.0], [2007.0, 1995.0], [2007.0, 1996.0], [2007.0, 1997.0], [2007.0, 1998.0], [2007.0, 1999.0], [2007.0, 2000.0], [2007.0, 2001.0], [2007.0, 2002.0], [2007.0, 2003.0], [2007.0, 2004.0], [2007.0, 2005.0], [2007.0, 2006.0], [2008.0, 1979.0], [2008.0, 1985.0], [2008.0, 1988.0], [2008.0, 1992.0], [2008.0, 1994.0], [2008.0, 1995.0], [2008.0, 1996.0], [2008.0, 1997.0], [2008.0, 1998.0], [2008.0, 1999.0], [2008.0, 2000.0], [2008.0, 2001.0], [2008.0, 2002.0], [2008.0, 2003.0], [2008.0, 2004.0], [2008.0, 2005.0], [2008.0, 2006.0], [2008.0, 2007.0], [2009.0, 1979.0], [2009.0, 1985.0], [2009.0, 1988.0], [2009.0, 1992.0], [2009.0, 1994.0], [2009.0, 1995.0], [2009.0, 1996.0], [2009.0, 1997.0], [2009.0, 1998.0], [2009.0, 1999.0], [2009.0, 2000.0], [2009.0, 2001.0], [2009.0, 2002.0], [2009.0, 2003.0], [2009.0, 2004.0], [2009.0, 2005.0], [2009.0, 2006.0], [2009.0, 2007.0], [2009.0, 2008.0], [2010.0, 1979.0], [2010.0, 1985.0], [2010.0, 1988.0], [2010.0, 1992.0], [2010.0, 1994.0], [2010.0, 1995.0], [2010.0, 1996.0], [2010.0, 1997.0], [2010.0, 1998.0], [2010.0, 1999.0], [2010.0, 2000.0], [2010.0, 2001.0], [2010.0, 2002.0], [2010.0, 2003.0], [2010.0, 2004.0], [2010.0, 2006.0], [2010.0, 2007.0], [2010.0, 2008.0], [2010.0, 2009.0], [2011.0, 1979.0], [2011.0, 1985.0], [2011.0, 1988.0], [2011.0, 1992.0], [2011.0, 1994.0], [2011.0, 1995.0], [2011.0, 1996.0], [2011.0, 1997.0], [2011.0, 1998.0], [2011.0, 1999.0], [2011.0, 2000.0], [2011.0, 2001.0], [2011.0, 2002.0], [2011.0, 2003.0], [2011.0, 2004.0], [2011.0, 2005.0], [2011.0, 2006.0], [2011.0, 2007.0], [2011.0, 2008.0], [2011.0, 2009.0], [2011.0, 2010.0], [2012.0, 1979.0], [2012.0, 1985.0], [2012.0, 1988.0], [2012.0, 1992.0], [2012.0, 1994.0], [2012.0, 1995.0], [2012.0, 1996.0], [2012.0, 1997.0], [2012.0, 1998.0], [2012.0, 1999.0], [2012.0, 2000.0], [2012.0, 2001.0], [2012.0, 2002.0], [2012.0, 2003.0], [2012.0, 2004.0], [2012.0, 2005.0], [2012.0, 2006.0], [2012.0, 2007.0], [2012.0, 2008.0], [2012.0, 2009.0], [2012.0, 2010.0], [2012.0, 2011.0], [2013.0, 1979.0], [2013.0, 1985.0], [2013.0, 1988.0], [2013.0, 1992.0], [2013.0, 1994.0], [2013.0, 1995.0], [2013.0, 1996.0], [2013.0, 1997.0], [2013.0, 1998.0], [2013.0, 1999.0], [2013.0, 2000.0], [2013.0, 2001.0], [2013.0, 2002.0], [2013.0, 2003.0], [2013.0, 2004.0], [2013.0, 2006.0], [2013.0, 2007.0], [2013.0, 2008.0], [2013.0, 2009.0], [2013.0, 2010.0], [2013.0, 2011.0], [2013.0, 2012.0], [2014.0, 1979.0], [2014.0, 1985.0], [2014.0, 1988.0], [2014.0, 1992.0], [2014.0, 1994.0], [2014.0, 1995.0], [2014.0, 1996.0], [2014.0, 1997.0], [2014.0, 1998.0], [2014.0, 1999.0], [2014.0, 2000.0], [2014.0, 2001.0], [2014.0, 2002.0], [2014.0, 2003.0], [2014.0, 2004.0], [2014.0, 2006.0], [2014.0, 2007.0], [2014.0, 2008.0], [2014.0, 2009.0], [2014.0, 2010.0], [2014.0, 2011.0], [2014.0, 2012.0], [2014.0, 2013.0], [2015.0, 1979.0], [2015.0, 1985.0], [2015.0, 1988.0], [2015.0, 1992.0], [2015.0, 1994.0], [2015.0, 1995.0], [2015.0, 1996.0], [2015.0, 1997.0], [2015.0, 1998.0], [2015.0, 1999.0], [2015.0, 2000.0], [2015.0, 2001.0], [2015.0, 2002.0], [2015.0, 2003.0], [2015.0, 2004.0], [2015.0, 2006.0], [2015.0, 2007.0], [2015.0, 2008.0], [2015.0, 2009.0], [2015.0, 2010.0], [2015.0, 2011.0], [2015.0, 2012.0], [2015.0, 2013.0], [2015.0, 2014.0], [2016.0, 1979.0], [2016.0, 1985.0], [2016.0, 1988.0], [2016.0, 1992.0], [2016.0, 1994.0], [2016.0, 1995.0], [2016.0, 1996.0], [2016.0, 1997.0], [2016.0, 1998.0], [2016.0, 1999.0], [2016.0, 2000.0], [2016.0, 2001.0], [2016.0, 2002.0], [2016.0, 2003.0], [2016.0, 2004.0], [2016.0, 2006.0], [2016.0, 2007.0], [2016.0, 2008.0], [2016.0, 2009.0], [2016.0, 2010.0], [2016.0, 2011.0], [2016.0, 2012.0], [2016.0, 2013.0], [2016.0, 2014.0], [2016.0, 2015.0]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v_jA_YLFtk1q",
        "outputId": "e1bb5b99-1a32-48bf-946d-c7d4046dd554"
      },
      "source": [
        "(tYvsG_F,tYvsPt_F,tPlvsG_F,tPtvsG_F,tPtvsY_F,tGvsY_F,tGvsPt_F)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(F_onewayResult(statistic=5.207428950017051, pvalue=1.886641365255221e-12),\n",
              " F_onewayResult(statistic=0.7250892559971256, pvalue=0.8074425391863829),\n",
              " F_onewayResult(statistic=5.227169910500004, pvalue=1.8898946086109917e-09),\n",
              " F_onewayResult(statistic=13.366795170725425, pvalue=5.59593667524491e-05),\n",
              " F_onewayResult(statistic=46.02899609784036, pvalue=6.286438377196204e-14),\n",
              " F_onewayResult(statistic=16.25881634351637, pvalue=5.85510791912265e-23),\n",
              " F_onewayResult(statistic=0.6008021488497898, pvalue=0.796381697288864))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5esOlx_VXzfo",
        "outputId": "d0184ba4-456d-497b-b6b8-75dc511cb333"
      },
      "source": [
        "(PlvsRS_F,GvsRS_F,YvsRS_F,PuvsRS_F,PtvsRS_F)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(F_onewayResult(statistic=6.5633466089567944, pvalue=0.00038018368161266017),\n",
              " F_onewayResult(statistic=8.726767944248904, pvalue=0.00011780318116774667),\n",
              " F_onewayResult(statistic=14.40933529650218, pvalue=2.4704696343599862e-08),\n",
              " F_onewayResult(statistic=4.309450197560952, pvalue=0.004868982218032098),\n",
              " F_onewayResult(statistic=0.8044330524679015, pvalue=0.5257444240219884))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dh6epdwvyg8z",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d27912e7-75ae-4fa6-906a-8ef6e48bd769"
      },
      "source": [
        "df20=df20.apply(standard)\n",
        "PlvsRS_F=f_oneway(df20['NA_Sales'],df20['EU_Sales'],df20['JP_Sales'],df20['Other_Sales'])\n",
        "df21=df21.apply(standard)\n",
        "\n",
        "GvsRS_F=f_oneway(df21['NA_Sales'],df21['EU_Sales'],df21['JP_Sales'],df21['Other_Sales'])\n",
        "df22=df22.apply(standard)\n",
        "\n",
        "YvsRS_F=f_oneway(df22['NA_Sales'],df22['EU_Sales'],df22['JP_Sales'],df22['Other_Sales'])\n",
        "df23=df23.apply(standard)\n",
        "\n",
        "PuvsRS_F=f_oneway(df23['NA_Sales'],df23['EU_Sales'],df23['JP_Sales'],df23['Other_Sales'])\n",
        "df24=df24.apply(standard)\n",
        "\n",
        "PtvsRS_F=f_oneway(df24['NA_Sales'],df24['EU_Sales'],df24['JP_Sales'],df24['Other_Sales'])\n",
        "df2=df2.apply(standard)\n",
        "df5=df5.apply(standard)\n",
        "df7=df7.apply(standard)\n",
        "df11=df11.apply(standard)\n",
        "df17=df17.apply(standard)\n",
        "\n",
        "(PlvsRS_F,GvsRS_F,YvsRS_F,PuvsRS_F,PtvsRS_F)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(F_onewayResult(statistic=1.504561323264268e-30, pvalue=1.0),\n",
              " F_onewayResult(statistic=1.8043823656747968e-31, pvalue=1.0),\n",
              " F_onewayResult(statistic=4.253275047316622e-30, pvalue=1.0),\n",
              " F_onewayResult(statistic=1.8607126033385713e-28, pvalue=1.0),\n",
              " F_onewayResult(statistic=2.0200865194461676e-32, pvalue=1.0))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a_GWSGhaTZNo",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 432
        },
        "outputId": "85fc8217-908d-4e2f-9d34-5e8fac3ae581"
      },
      "source": [
        "#출고량 100만장 이상을 기준으로\n",
        "df1=df.loc[df['Sales_sum']>=1]\n",
        "\n",
        "df25=pd.crosstab(df1['Year'],df1['Platform'])#\n",
        "df26=pd.crosstab(df1['Year'],df1['Genre'])#\n",
        "df27=pd.crosstab(df1['Year'],df1['Publisher'])\n",
        "df28=pd.crosstab(df1['Year'],df1['Ptype'])#\n",
        "df29=pd.crosstab(df1['Genre'],df1['Platform'])#\n",
        "df30=pd.crosstab(df1['Genre'],df1['Publisher'])\n",
        "df31=pd.crosstab(df1['Genre'],df1['Ptype'])#PC console portable의 비율을 통해 포터블로 만들기 쉬운것 어려운것 대략적으로 알 수 있음\n",
        "df32=pd.crosstab(df1['Ptype'],df1['Publisher'])\n",
        "df33=pd.crosstab(df1['Platform'],df1['Publisher'])\n",
        "df21"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>NA_Sales</th>\n",
              "      <th>EU_Sales</th>\n",
              "      <th>JP_Sales</th>\n",
              "      <th>Other_Sales</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Genre</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Sports</th>\n",
              "      <td>690.51</td>\n",
              "      <td>386.47</td>\n",
              "      <td>135.37</td>\n",
              "      <td>144.66</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Platform</th>\n",
              "      <td>446.44</td>\n",
              "      <td>201.61</td>\n",
              "      <td>130.65</td>\n",
              "      <td>51.53</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Racing</th>\n",
              "      <td>357.19</td>\n",
              "      <td>237.10</td>\n",
              "      <td>56.40</td>\n",
              "      <td>87.09</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Role-Playing</th>\n",
              "      <td>319.54</td>\n",
              "      <td>183.58</td>\n",
              "      <td>344.95</td>\n",
              "      <td>86.93</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Puzzle</th>\n",
              "      <td>133.34</td>\n",
              "      <td>50.73</td>\n",
              "      <td>56.87</td>\n",
              "      <td>12.52</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Misc</th>\n",
              "      <td>428.66</td>\n",
              "      <td>225.48</td>\n",
              "      <td>128.14</td>\n",
              "      <td>89.57</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Shooter</th>\n",
              "      <td>581.92</td>\n",
              "      <td>323.04</td>\n",
              "      <td>38.28</td>\n",
              "      <td>102.65</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Simulation</th>\n",
              "      <td>182.96</td>\n",
              "      <td>113.35</td>\n",
              "      <td>63.70</td>\n",
              "      <td>51.46</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Action</th>\n",
              "      <td>874.81</td>\n",
              "      <td>532.15</td>\n",
              "      <td>167.89</td>\n",
              "      <td>196.44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Fighting</th>\n",
              "      <td>223.20</td>\n",
              "      <td>101.22</td>\n",
              "      <td>87.34</td>\n",
              "      <td>46.63</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Adventure</th>\n",
              "      <td>105.66</td>\n",
              "      <td>74.04</td>\n",
              "      <td>61.94</td>\n",
              "      <td>16.81</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Strategy</th>\n",
              "      <td>68.70</td>\n",
              "      <td>55.31</td>\n",
              "      <td>49.46</td>\n",
              "      <td>11.36</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              NA_Sales  EU_Sales  JP_Sales  Other_Sales\n",
              "Genre                                                  \n",
              "Sports          690.51    386.47    135.37       144.66\n",
              "Platform        446.44    201.61    130.65        51.53\n",
              "Racing          357.19    237.10     56.40        87.09\n",
              "Role-Playing    319.54    183.58    344.95        86.93\n",
              "Puzzle          133.34     50.73     56.87        12.52\n",
              "Misc            428.66    225.48    128.14        89.57\n",
              "Shooter         581.92    323.04     38.28       102.65\n",
              "Simulation      182.96    113.35     63.70        51.46\n",
              "Action          874.81    532.15    167.89       196.44\n",
              "Fighting        223.20    101.22     87.34        46.63\n",
              "Adventure       105.66     74.04     61.94        16.81\n",
              "Strategy         68.70     55.31     49.46        11.36"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "of-J9iIskjda",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 611
        },
        "outputId": "4dd35484-0909-41c8-cd6b-f60009174050"
      },
      "source": [
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "a4_dims = (15, 10)\n",
        "fig, ax = plt.subplots(figsize=a4_dims)\n",
        "ax = sns.heatmap(df2,xticklabels=1,yticklabels=1)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EBjCszIW5Jd2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 402
        },
        "outputId": "1969e749-d63f-4ccf-c497-3c34e56f1a27"
      },
      "source": [
        "df1=indsum(df1.loc[:,['Publisher','Sales_sum']])\n",
        "df1.columns=['pu','Sales_sum']\n",
        "df1=df1.sort_values(by='Sales_sum' ,ascending=False)\n",
        "df1"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>pu</th>\n",
              "      <th>Sales_sum</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Nintendo</td>\n",
              "      <td>1654.18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>759.45</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Activision</td>\n",
              "      <td>499.30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Sony Computer Entertainment</td>\n",
              "      <td>456.72</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Take-Two Interactive</td>\n",
              "      <td>306.91</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>GSP</td>\n",
              "      <td>1.07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>100</th>\n",
              "      <td>Play It</td>\n",
              "      <td>1.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>101</th>\n",
              "      <td>Slightly Mad Studios</td>\n",
              "      <td>1.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>102</th>\n",
              "      <td>Tomy Corporation</td>\n",
              "      <td>1.02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>103</th>\n",
              "      <td>Sammy Corporation</td>\n",
              "      <td>1.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>104 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                              pu  Sales_sum\n",
              "0                       Nintendo    1654.18\n",
              "8                Electronic Arts     759.45\n",
              "4                     Activision     499.30\n",
              "3    Sony Computer Entertainment     456.72\n",
              "2           Take-Two Interactive     306.91\n",
              "..                           ...        ...\n",
              "99                           GSP       1.07\n",
              "100                      Play It       1.06\n",
              "101         Slightly Mad Studios       1.03\n",
              "102             Tomy Corporation       1.02\n",
              "103            Sammy Corporation       1.00\n",
              "\n",
              "[104 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 101
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1q6iV6OmP9K_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 846
        },
        "outputId": "cddfd115-f6a7-46b4-8165-9a1e954a45c4"
      },
      "source": [
        "df.loc[df['Publisher']=='Electronic Arts'].loc[df['Platform']=='PS4']\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Name</th>\n",
              "      <th>Platform</th>\n",
              "      <th>Year</th>\n",
              "      <th>Genre</th>\n",
              "      <th>Publisher</th>\n",
              "      <th>NA_Sales</th>\n",
              "      <th>EU_Sales</th>\n",
              "      <th>JP_Sales</th>\n",
              "      <th>Other_Sales</th>\n",
              "      <th>Sales_sum</th>\n",
              "      <th>Ptype</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>index</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>FIFA 16</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>1.11</td>\n",
              "      <td>6.06</td>\n",
              "      <td>0.06</td>\n",
              "      <td>1.26</td>\n",
              "      <td>8.49</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>112</th>\n",
              "      <td>Star Wars Battlefront (2015)</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Shooter</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>2.93</td>\n",
              "      <td>3.29</td>\n",
              "      <td>0.22</td>\n",
              "      <td>1.23</td>\n",
              "      <td>7.67</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>144</th>\n",
              "      <td>FIFA 15</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.79</td>\n",
              "      <td>4.29</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1.47</td>\n",
              "      <td>6.60</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>241</th>\n",
              "      <td>FIFA 17</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.28</td>\n",
              "      <td>3.75</td>\n",
              "      <td>0.06</td>\n",
              "      <td>0.69</td>\n",
              "      <td>4.78</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>406</th>\n",
              "      <td>Battlefield 4</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2013.0</td>\n",
              "      <td>Shooter</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>1.34</td>\n",
              "      <td>1.54</td>\n",
              "      <td>0.17</td>\n",
              "      <td>0.38</td>\n",
              "      <td>3.43</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>438</th>\n",
              "      <td>Madden NFL 16</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>2.33</td>\n",
              "      <td>0.30</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.59</td>\n",
              "      <td>3.22</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>516</th>\n",
              "      <td>FIFA 14</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2013.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.61</td>\n",
              "      <td>1.85</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.33</td>\n",
              "      <td>2.90</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>675</th>\n",
              "      <td>Need for Speed (2015)</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Racing</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.49</td>\n",
              "      <td>1.49</td>\n",
              "      <td>0.05</td>\n",
              "      <td>0.36</td>\n",
              "      <td>2.39</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>815</th>\n",
              "      <td>Battlefield: Hardline</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Shooter</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.70</td>\n",
              "      <td>0.93</td>\n",
              "      <td>0.14</td>\n",
              "      <td>0.32</td>\n",
              "      <td>2.09</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>857</th>\n",
              "      <td>Madden NFL 15</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>1.53</td>\n",
              "      <td>0.25</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.25</td>\n",
              "      <td>2.03</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>876</th>\n",
              "      <td>Dragon Age: Inquisition</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Role-Playing</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.72</td>\n",
              "      <td>0.86</td>\n",
              "      <td>0.08</td>\n",
              "      <td>0.32</td>\n",
              "      <td>1.98</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>929</th>\n",
              "      <td>Need for Speed Rivals</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2013.0</td>\n",
              "      <td>Racing</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.73</td>\n",
              "      <td>0.91</td>\n",
              "      <td>0.03</td>\n",
              "      <td>0.22</td>\n",
              "      <td>1.89</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1644</th>\n",
              "      <td>Madden NFL 17</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.92</td>\n",
              "      <td>0.08</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.23</td>\n",
              "      <td>1.23</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2018</th>\n",
              "      <td>EA Sports UFC</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.46</td>\n",
              "      <td>0.40</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.17</td>\n",
              "      <td>1.04</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2351</th>\n",
              "      <td>EA Sports UFC 2</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.28</td>\n",
              "      <td>0.47</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.14</td>\n",
              "      <td>0.89</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2527</th>\n",
              "      <td>Madden NFL 25</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2013.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.62</td>\n",
              "      <td>0.13</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.07</td>\n",
              "      <td>0.82</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3295</th>\n",
              "      <td>Plants vs. Zombies: Garden Warfare</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Shooter</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.21</td>\n",
              "      <td>0.29</td>\n",
              "      <td>0.01</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.62</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3335</th>\n",
              "      <td>NHL 16</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.35</td>\n",
              "      <td>0.15</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.11</td>\n",
              "      <td>0.61</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3513</th>\n",
              "      <td>NHL 15</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.30</td>\n",
              "      <td>0.18</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.58</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3983</th>\n",
              "      <td>Rory McIlroy PGA Tour</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Action</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.18</td>\n",
              "      <td>0.24</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.08</td>\n",
              "      <td>0.50</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4388</th>\n",
              "      <td>Plants vs. Zombies: Garden Warfare 2</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Shooter</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.17</td>\n",
              "      <td>0.21</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.07</td>\n",
              "      <td>0.45</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5642</th>\n",
              "      <td>Mirror's Edge Catalyst</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Platform</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.14</td>\n",
              "      <td>0.02</td>\n",
              "      <td>0.06</td>\n",
              "      <td>0.32</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7041</th>\n",
              "      <td>NBA Live 15</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2014.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.15</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.23</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7291</th>\n",
              "      <td>NBA Live 14</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2013.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.14</td>\n",
              "      <td>0.05</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.03</td>\n",
              "      <td>0.22</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7352</th>\n",
              "      <td>NHL 17</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2016.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.13</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.21</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8386</th>\n",
              "      <td>NBA Live 16</td>\n",
              "      <td>PS4</td>\n",
              "      <td>2015.0</td>\n",
              "      <td>Sports</td>\n",
              "      <td>Electronic Arts</td>\n",
              "      <td>0.10</td>\n",
              "      <td>0.04</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.03</td>\n",
              "      <td>0.17</td>\n",
              "      <td>console</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                       Name Platform  ...  Sales_sum    Ptype\n",
              "index                                                 ...                    \n",
              "98                                  FIFA 16      PS4  ...       8.49  console\n",
              "112            Star Wars Battlefront (2015)      PS4  ...       7.67  console\n",
              "144                                 FIFA 15      PS4  ...       6.60  console\n",
              "241                                 FIFA 17      PS4  ...       4.78  console\n",
              "406                           Battlefield 4      PS4  ...       3.43  console\n",
              "438                           Madden NFL 16      PS4  ...       3.22  console\n",
              "516                                 FIFA 14      PS4  ...       2.90  console\n",
              "675                   Need for Speed (2015)      PS4  ...       2.39  console\n",
              "815                   Battlefield: Hardline      PS4  ...       2.09  console\n",
              "857                           Madden NFL 15      PS4  ...       2.03  console\n",
              "876                 Dragon Age: Inquisition      PS4  ...       1.98  console\n",
              "929                   Need for Speed Rivals      PS4  ...       1.89  console\n",
              "1644                          Madden NFL 17      PS4  ...       1.23  console\n",
              "2018                          EA Sports UFC      PS4  ...       1.04  console\n",
              "2351                        EA Sports UFC 2      PS4  ...       0.89  console\n",
              "2527                          Madden NFL 25      PS4  ...       0.82  console\n",
              "3295     Plants vs. Zombies: Garden Warfare      PS4  ...       0.62  console\n",
              "3335                                 NHL 16      PS4  ...       0.61  console\n",
              "3513                                 NHL 15      PS4  ...       0.58  console\n",
              "3983                  Rory McIlroy PGA Tour      PS4  ...       0.50  console\n",
              "4388   Plants vs. Zombies: Garden Warfare 2      PS4  ...       0.45  console\n",
              "5642                 Mirror's Edge Catalyst      PS4  ...       0.32  console\n",
              "7041                            NBA Live 15      PS4  ...       0.23  console\n",
              "7291                            NBA Live 14      PS4  ...       0.22  console\n",
              "7352                                 NHL 17      PS4  ...       0.21  console\n",
              "8386                            NBA Live 16      PS4  ...       0.17  console\n",
              "\n",
              "[26 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 107
        }
      ]
    }
  ]
}