{"id":20081,"date":"2026-01-16T16:24:26","date_gmt":"2026-01-16T08:24:26","guid":{"rendered":"https:\/\/92it.top\/?p=20081"},"modified":"2026-01-16T16:24:26","modified_gmt":"2026-01-16T08:24:26","slug":"%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0_%e7%89%b9%e5%be%81%e5%b7%a5%e7%a8%8b","status":"publish","type":"post","link":"https:\/\/92it.top\/?p=20081","title":{"rendered":"\u673a\u5668\u5b66\u4e60_\u7279\u5f81\u5de5\u7a0b"},"content":{"rendered":"\n

\u524d\u8a00\u200b\ud83d\udd16<\/strong><\/p>\n\n\n\n

\n\n\n\n

\u60f3\u8c61\u4e00\u4e0b\uff0c\u4f60\u662f\u4e00\u4f4d\u53a8\u5e08\uff0c\u8981\u505a\u4e00\u9053\u7f8e\u5473\u7684\u83dc\u80b4\uff0c\u673a\u5668\u5b66\u4e60\u6a21\u578b\u5c31\u50cf\u4f60\u7684”\u70f9\u996a\u7b97\u6cd5”\uff0c\u800c\u539f\u59cb\u6570\u636e\u5c31\u662f\u4f60\u4ece\u5e02\u573a\u4e70\u56de\u6765\u7684\u5404\u79cd\u98df\u6750\uff1a\u6709\u852c\u83dc\u3001\u8089\u7c7b\u3001\u8c03\u6599\uff0c\u4f46\u53ef\u80fd\u6709\u4e9b\u662f\u5e26\u6ce5\u7684\uff0c\u6709\u4e9b\u662f\u6574\u5757\u7684\uff0c\u6709\u4e9b\u5473\u9053\u5f88\u51b2\u3002\u7279\u5f81\u5de5\u7a0b<\/strong>\uff0c\u5c31\u662f\u5c06\u8fd9\u4e9b\u539f\u59cb”\u98df\u6750”\u8fdb\u884c\u6e05\u6d17\u3001\u5207\u5272\u3001\u814c\u5236\u3001\u642d\u914d\uff0c\u6700\u7ec8\u5904\u7406\u6210\u53ef\u4ee5\u76f4\u63a5\u4e0b\u9505\u70f9\u996a\u7684”\u534a\u6210\u54c1”\u7684\u8fc7\u7a0b\uff0c\u5b83\u662f\u8fde\u63a5\u539f\u59cb\u6570\u636e\u4e0e\u673a\u5668\u5b66\u4e60\u6a21\u578b\u7684\u6865\u6881\uff0c\u662f\u51b3\u5b9a\u6a21\u578b\u6027\u80fd\u4e0a\u9650\u7684\u5173\u952e\u6b65\u9aa4\u3002<\/p>\n\n\n\n

\u7b80\u5355\u6765\u8bf4\uff0c\u7279\u5f81\u5de5\u7a0b<\/strong>\u662f\u5229\u7528\u9886\u57df\u77e5\u8bc6\uff0c\u901a\u8fc7\u4e00\u7cfb\u5217\u6280\u672f\u624b\u6bb5\uff0c\u4ece\u539f\u59cb\u6570\u636e\u4e2d\u63d0\u53d6\u3001\u6784\u9020\u3001\u9009\u62e9\u51fa\u5bf9\u673a\u5668\u5b66\u4e60\u6a21\u578b\u66f4\u6709\u4ef7\u503c\u3001\u66f4\u6613\u5b66\u4e60\u7684\u7279\u5f81\uff08\u53d8\u91cf\uff09\u7684\u8fc7\u7a0b\u3002<\/p>\n\n\n\n

\u3000\u3000<\/p>\n\n\n\n

\u4e3a\u4ec0\u4e48\u7279\u5f81\u5de5\u7a0b\u5982\u6b64\u91cd\u8981\uff1f\ud83d\udd16<\/strong><\/p>\n\n\n\n

\n\n\n\n

\u5728\u673a\u5668\u5b66\u4e60\u9879\u76ee\u4e2d\uff0c\u6570\u636e\u548c\u7279\u5f81\u7684\u8d28\u91cf\u76f4\u63a5\u51b3\u5b9a\u4e86\u6a21\u578b\u6027\u80fd\u7684\u4e0a\u9650\uff0c\u800c\u6a21\u578b\u548c\u7b97\u6cd5\u53ea\u662f\u903c\u8fd1\u8fd9\u4e2a\u4e0a\u9650\u3002\u4e00\u4e2a\u4f18\u79c0\u7684\u7279\u5f81\u5de5\u7a0b\u53ef\u4ee5\uff1a<\/p>\n\n\n\n

    \n
  1. \u63d0\u5347\u6a21\u578b\u6027\u80fd<\/strong>\uff1a\u597d\u7684\u7279\u5f81\u80fd\u8ba9\u6a21\u578b\u66f4\u5bb9\u6613\u53d1\u73b0\u6570\u636e\u4e2d\u7684\u89c4\u5f8b\u3002<\/li>\n\n\n\n
  2. \u52a0\u901f\u6a21\u578b\u8bad\u7ec3<\/strong>\uff1a\u51cf\u5c11\u65e0\u5173\u6216\u5197\u4f59\u7279\u5f81\uff0c\u53ef\u4ee5\u964d\u4f4e\u8ba1\u7b97\u590d\u6742\u5ea6\u3002<\/li>\n\n\n\n
  3. \u589e\u5f3a\u6a21\u578b\u6cdb\u5316\u80fd\u529b<\/strong>\uff1a\u9632\u6b62\u6a21\u578b\u8fc7\u62df\u5408\u4e8e\u8bad\u7ec3\u6570\u636e\u4e2d\u7684\u566a\u58f0\u3002<\/li>\n\n\n\n
  4. \u9002\u5e94\u6a21\u578b\u9700\u6c42<\/strong>\uff1a\u4e0d\u540c\u7684\u6a21\u578b\u5bf9\u6570\u636e\u6709\u4e0d\u540c\u7684\u5047\u8bbe\uff08\u5982\u7ebf\u6027\u6a21\u578b\u5047\u8bbe\u7ebf\u6027\u5173\u7cfb\uff09\uff0c\u7279\u5f81\u5de5\u7a0b\u53ef\u4ee5\u4f7f\u6570\u636e\u6ee1\u8db3\u8fd9\u4e9b\u5047\u8bbe\u3002<\/li>\n<\/ol>\n\n\n\n

    \u6211\u4eec\u53ef\u4ee5\u7528\u4ee5\u4e0b\u6d41\u7a0b\u56fe\u6765\u76f4\u89c2\u7406\u89e3\u7279\u5f81\u5de5\u7a0b\u5728\u6574\u4e2a\u673a\u5668\u5b66\u4e60\u6d41\u7a0b\u4e2d\u7684\u4f4d\u7f6e\uff1a<\/p>\n\n\n

    \n
    \"\"<\/figure><\/div>\n\n\n

    \u3000\u3000<\/p>\n\n\n\n

    \u7279\u5f81\u5de5\u7a0b\u7684\u6838\u5fc3\u64cd\u4f5c\ud83d\udd16<\/strong><\/p>\n\n\n\n

    \n\n\n\n

    \u7279\u5f81\u5de5\u7a0b\u4e3b\u8981\u5305\u542b\u4e09\u5927\u7c7b\u64cd\u4f5c\uff1a\u7279\u5f81\u5904\u7406<\/strong>\u3001\u7279\u5f81\u6784\u9020<\/strong>\u548c\u7279\u5f81\u9009\u62e9<\/strong>\u3002<\/p>\n\n\n\n

    \ud83d\udd391. \u7279\u5f81\u5904\u7406<\/strong><\/p>\n\n\n\n

    \u8fd9\u662f\u6700\u57fa\u7840\u7684\u4e00\u6b65\uff0c\u76ee\u7684\u662f\u5c06\u539f\u59cb\u6570\u636e”\u6e05\u6d17”\u6210\u5e72\u51c0\u3001\u89c4\u6574\u7684\u683c\u5f0f\u3002<\/p>\n\n\n\n

      \n
    • a) \u5904\u7406\u7f3a\u5931\u503c<\/li>\n<\/ul>\n\n\n\n

      \u6570\u636e\u4e2d\u7ecf\u5e38\u5b58\u5728\u7f3a\u5931\u503c\uff08\u5982 NaN<\/code>, NULL<\/code>\uff09\uff0c\u9700\u8981\u5408\u7406\u5904\u7406\u3002<\/p>\n\n\n\n

      \u7406\u65b9\u6cd5<\/th>\u8bf4\u660e<\/th>\u9002\u7528\u573a\u666f<\/th><\/tr><\/thead>
      \u5220\u9664<\/strong><\/td>\u76f4\u63a5\u5220\u9664\u7f3a\u5931\u503c\u6240\u5728\u7684\u884c\u6216\u5217<\/td>\u7f3a\u5931\u6570\u636e\u6781\u5c11\uff0c\u6216\u8be5\u7279\u5f81\u4e0d\u91cd\u8981\u65f6<\/td><\/tr>
      \u586b\u5145<\/strong><\/td>\u7528\u67d0\u4e2a\u503c\u586b\u5145\uff0c\u5982\u5747\u503c\u3001\u4e2d\u4f4d\u6570\u3001\u4f17\u6570\u6216\u4e00\u4e2a\u7279\u6b8a\u503c\uff08\u5982 -1\uff09<\/td>\u6700\u5e38\u7528\u7684\u65b9\u6cd5\uff0c\u9002\u7528\u4e8e\u5404\u79cd\u60c5\u51b5<\/td><\/tr>
      \u63d2\u503c<\/strong><\/td>\u7528\u65f6\u95f4\u5e8f\u5217\u6216\u76f8\u90bb\u6570\u636e\u70b9\u8fdb\u884c\u63d2\u503c\u8ba1\u7b97<\/td>\u65f6\u95f4\u5e8f\u5217\u6570\u636e<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

      \u3000\u3000<\/p>\n\n\n\n

      \u4ee3\u7801\u793a\u4f8b\uff08\u4f7f\u7528 Python \u7684 pandas \u5e93\uff09\uff1a<\/p>\n\n\n\n

      import pandas as pd\nimport numpy as np\n\n# \u521b\u5efa\u4e00\u4e2a\u5305\u542b\u7f3a\u5931\u503c\u7684\u793a\u4f8b DataFrame\ndata = {'\u5e74\u9f84': [25, np.nan, 30, 35, np.nan],\n        '\u5de5\u8d44': [50000, 54000, np.nan, 62000, 58000],\n        '\u57ce\u5e02': ['\u5317\u4eac', '\u4e0a\u6d77', '\u5e7f\u5dde', np.nan, '\u5317\u4eac']}\ndf = pd.DataFrame(data)\nprint(\"\u539f\u59cb\u6570\u636e\uff1a\")\nprint(df)\n\n# 1. \u5220\u9664\u7f3a\u5931\u503c\uff08\u5220\u9664\u4efb\u4f55\u5305\u542b NaN \u7684\u884c\uff09\ndf_dropped = df.dropna()\nprint(\"\\n\u5220\u9664\u7f3a\u5931\u503c\u540e\uff1a\")\nprint(df_dropped)\n\n# 2. \u586b\u5145\u7f3a\u5931\u503c\n# \u6570\u503c\u5217\u7528\u5747\u503c\u586b\u5145\ndf_filled = df.copy()\ndf_filled['\u5e74\u9f84'].fillna(df_filled['\u5e74\u9f84'].mean(), inplace=True)\ndf_filled['\u5de5\u8d44'].fillna(df_filled['\u5de5\u8d44'].mean(), inplace=True)\n# \u7c7b\u522b\u5217\u7528\u4f17\u6570\u586b\u5145\ndf_filled['\u57ce\u5e02'].fillna(df_filled['\u57ce\u5e02'].mode()[0], inplace=True)\nprint(\"\\n\u586b\u5145\u7f3a\u5931\u503c\u540e\uff1a\")\nprint(df_filled)<\/pre>\n\n\n
      \n
      \"\"<\/figure><\/div>\n\n\n
      \u3000\u3000<\/p>\n\n\n\n
        \n
      • b) \u5904\u7406\u5f02\u5e38\u503c<\/li>\n<\/ul>\n\n\n\n
        \u5f02\u5e38\u503c\u662f\u4e0e\u5927\u90e8\u5206\u6570\u636e\u660e\u663e\u4e0d\u540c\u7684\u503c\uff0c\u53ef\u80fd\u4f1a\u5e72\u6270\u6a21\u578b\u3002\u5e38\u7528\u68c0\u6d4b\u65b9\u6cd5\u6709\uff1a<\/p>\n\n\n\n
          \n
        • \u6807\u51c6\u5dee\u6cd5<\/strong>\uff1a\u8ba4\u4e3a\u8d85\u51fa\u5747\u503c \u00b1 3\u500d\u6807\u51c6\u5dee\u8303\u56f4\u7684\u503c\u662f\u5f02\u5e38\u503c\u3002<\/li>\n\n\n\n
        • \u7bb1\u7ebf\u56fe\u6cd5<\/strong>\uff1a\u8ba4\u4e3a\u5c0f\u4e8e\u00a0Q1 - 1.5*IQR<\/code>\u00a0\u6216\u5927\u4e8e\u00a0Q3 + 1.5*IQR<\/code>\u00a0\u7684\u503c\u662f\u5f02\u5e38\u503c\uff08IQR = Q3 - Q1<\/code>\uff09\u3002<\/li>\n<\/ul>\n\n\n\n
          \u5904\u7406\u65b9\u5f0f\u53ef\u4ee5\u662f\u5220\u9664\u3001\u66ff\u6362\u4e3a\u8fb9\u754c\u503c\u6216\u89c6\u4e3a\u7f3a\u5931\u503c\u5904\u7406\u3002<\/p>\n\n\n\n
          \u3000\u3000<\/p>\n\n\n\n
            \n
          • c) \u6570\u636e\u6807\u51c6\u5316\/\u5f52\u4e00\u5316<\/li>\n<\/ul>\n\n\n\n
            \u8bb8\u591a\u6a21\u578b\uff08\u5982 SVM\u3001KNN\u3001\u795e\u7ecf\u7f51\u7edc\uff09\u5bf9\u7279\u5f81\u7684\u5c3a\u5ea6\u654f\u611f\u3002\u6211\u4eec\u9700\u8981\u5c06\u4e0d\u540c\u5c3a\u5ea6\u7684\u7279\u5f81\u8f6c\u6362\u5230\u540c\u4e00\u5c3a\u5ea6\u3002<\/p>\n\n\n\n

            \u516c\u5f0f<\/th>            		\u8bf4\u660e<\/th>\u9002\u7528\u573a\u666f<\/th><\/tr><\/thead>
            \u6807\u51c6\u5316<\/strong><\/td>(x - \u5747\u503c) \/ \u6807\u51c6\u5dee<\/code><\/td>\u5904\u7406\u540e\u6570\u636e\u5747\u503c\u4e3a0\uff0c\u6807\u51c6\u5dee\u4e3a1<\/td>\u6570\u636e\u5206\u5e03\u8fd1\u4f3c\u6b63\u6001\u5206\u5e03\u65f6<\/td><\/tr>
            \u5f52\u4e00\u5316<\/strong><\/td>(x - min) \/ (max - min)<\/code><\/td>\u5c06\u6570\u636e\u7f29\u653e\u5230 [0, 1] \u533a\u95f4<\/td>\u6570\u636e\u8fb9\u754c\u660e\u786e\uff0c\u9700\u8981\u5feb\u901f\u8ba1\u7b97\u65f6<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n
            \u3000\u3000<\/p>\n\n\n\n
            \u4ee3\u7801\u793a\u4f8b\uff08\u4f7f\u7528 scikit-learn \u5e93\uff09\uff1a<\/p>\n\n\n\n
            from sklearn.preprocessing import StandardScaler, MinMaxScaler\nimport numpy as np\n\n# \u793a\u4f8b\u6570\u636e\ndata = np.array([[1000, 25],\n                 [1500, 30],\n                 [800, 20],\n                 [1200, 28]])\n\n# \u6807\u51c6\u5316\nscaler_standard = StandardScaler()\ndata_standardized = scaler_standard.fit_transform(data)\nprint(\"\u6807\u51c6\u5316\u540e\u7684\u6570\u636e\uff08\u5747\u503c~0\uff0c \u6807\u51c6\u5dee~1\uff09\uff1a\")\nprint(data_standardized)\nprint(f\"\u5747\u503c\uff1a {data_standardized.mean(axis=0)}\")\nprint(f\"\u6807\u51c6\u5dee\uff1a {data_standardized.std(axis=0)}\")\n\n# \u5f52\u4e00\u5316\nscaler_minmax = MinMaxScaler()\ndata_normalized = scaler_minmax.fit_transform(data)\nprint(\"\\n\u5f52\u4e00\u5316\u540e\u7684\u6570\u636e\uff08\u8303\u56f4[0,1]\uff09\uff1a\")\nprint(data_normalized)<\/pre>\n\n\n
            \n
            \"\"<\/figure><\/div>\n\n\n
            \u3000\u3000<\/p>\n\n\n\n
            \ud83d\udd392. \u7279\u5f81\u6784\u9020<\/strong><\/p>\n\n\n\n
            \u901a\u8fc7\u7ec4\u5408\u6216\u8f6c\u6362\u73b0\u6709\u7279\u5f81\uff0c\u521b\u9020\u51fa\u65b0\u7684\u3001\u66f4\u5177\u9884\u6d4b\u529b\u7684\u7279\u5f81\u3002<\/p>\n\n\n\n
              \n
            • a) \u5bf9\u6570\u503c\u7279\u5f81\u8fdb\u884c\u53d8\u6362<\/li>\n<\/ul>\n\n\n\n
                \n
              • \u591a\u9879\u5f0f\u7279\u5f81<\/strong>\uff1a\u521b\u5efa\u7279\u5f81\u7684\u5e73\u65b9\u3001\u7acb\u65b9\u7b49\uff0c\u5e2e\u52a9\u7ebf\u6027\u6a21\u578b\u5b66\u4e60\u975e\u7ebf\u6027\u5173\u7cfb\u3002<\/li>\n\n\n\n
              • \u5206\u7bb1<\/strong>\uff1a\u5c06\u8fde\u7eed\u5e74\u9f84\u5206\u4e3a”\u9752\u5e74”\u3001”\u4e2d\u5e74”\u3001”\u8001\u5e74”\u7b49\u533a\u95f4\uff0c\u5c06\u8fde\u7eed\u6570\u636e\u79bb\u6563\u5316\u3002<\/li>\n\n\n\n
              • \u6570\u5b66\u53d8\u6362<\/strong>\uff1a\u4f7f\u7528\u5bf9\u6570\u3001\u6307\u6570\u7b49\u53d8\u6362\u6539\u53d8\u6570\u636e\u5206\u5e03\u3002<\/li>\n<\/ul>\n\n\n\n