{"id":774,"date":"2024-07-11T08:22:35","date_gmt":"2024-07-11T08:22:35","guid":{"rendered":"http:\/\/www.ohsumin.com\/?p=774"},"modified":"2025-06-11T01:23:00","modified_gmt":"2025-06-11T01:23:00","slug":"%ec%82%ac%ed%9b%84%ed%99%95%eb%a5%a0","status":"publish","type":"post","link":"http:\/\/www.inmining.co.kr\/?p=774","title":{"rendered":"\uc0ac\ud6c4\ud655\ub960(Post Probability)~\ubca0\uc774\uc988\uc815\ub9ac(Baye’s theorem)"},"content":{"rendered":"\n

\uc0ac\ud6c4\ud655\ub960(Posterior Probability)\uc740 \ubca0\uc774\uc988 \uc815\ub9ac(Bayes’ theorem)\uc5d0 \uae30\ubc18\ud55c \uac1c\ub150\uc73c\ub85c, \ud2b9\uc815 \uc0ac\uac74\uc774 \ubc1c\uc0dd\ud55c \ud6c4\uc5d0 \uadf8 \uc0ac\uac74\uc5d0 \ub300\ud55c \uc870\uac74\ubd80 \ud655\ub960\uc744 \uc5c5\ub370\uc774\ud2b8\ud558\ub294 \uacfc\uc815\uc785\ub2c8\ub2e4. \uc774\ub294 \uc0c8\ub85c\uc6b4 \uc99d\uac70\ub098 \uc815\ubcf4\ub97c \uace0\ub824\ud558\uc5ec \uae30\uc874\uc758 \ud655\ub960\uc744 \uc218\uc815\ud558\ub294 \ubc29\ubc95\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n

\\(\\)<\/p>\n\n\n\n

<\/span>\ubca0\uc774\uc988 \uc815\ub9ac<\/span><\/h3>\n\n\n\n

\ubca0\uc774\uc988 \uc815\ub9ac\ub294 \ub2e4\uc74c\uacfc \uac19\uc740 \uc218\uc2dd\uc73c\ub85c \ud45c\ud604\ub429\ub2c8\ub2e4:<\/p>\n\n\n\n

\\[
P(A|B) = \\frac{P(B|A) \\cdot P(A)}{P(B)}
\\]<\/p>\n\n\n\n

\uc5ec\uae30\uc11c,<\/p>\n\n\n\n

    \n
  • \\( P(A|B) \\)\ub294 B\uac00 \ubc1c\uc0dd\ud55c \ud6c4 A\uac00 \ubc1c\uc0dd\ud560 \uc0ac\ud6c4\ud655\ub960\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
  • \\( P(B|A) \\)\ub294 A\uac00 \ubc1c\uc0dd\ud55c \uc870\uac74\uc5d0\uc11c B\uac00 \ubc1c\uc0dd\ud560 \ud655\ub960\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
  • \\( P(A) \\)\ub294 A\uac00 \ubc1c\uc0dd\ud560 \uc0ac\uc804\ud655\ub960(Prior Probability)\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
  • \\( P(B) \\)\ub294 B\uac00 \ubc1c\uc0dd\ud560 \uc804\uccb4 \ud655\ub960\uc785\ub2c8\ub2e4.<\/li>\n<\/ul>\n\n\n\n

    <\/span>\uc0ac\ud6c4\ud655\ub960\uc758 \ud574\uc11d<\/span><\/h3>\n\n\n\n

    \uc0ac\ud6c4\ud655\ub960\uc740 \uc0c8\ub85c\uc6b4 \uc99d\uac70\uac00 \uc8fc\uc5b4\uc84c\uc744 \ub54c \uc5b4\ub5a4 \uc0ac\uac74\uc774 \ubc1c\uc0dd\ud560 \ud655\ub960\uc744 \ub098\ud0c0\ub0c5\ub2c8\ub2e4. \uc774\ub294 \ub2e4\uc74c\uacfc \uac19\uc740 \ub2e8\uacc4\ub85c \uc774\ud574\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4:<\/p>\n\n\n\n

      \n
    1. \uc0ac\uc804\ud655\ub960(Prior Probability)<\/strong>: \uc0ac\uac74 A\uac00 \ubc1c\uc0dd\ud560 \ucd08\uae30 \ud655\ub960\uc744 \uc124\uc815\ud569\ub2c8\ub2e4. \uc774 \ud655\ub960\uc740 \uae30\uc874\uc758 \uc815\ubcf4\ub098 \uacbd\ud5d8\uc5d0 \uae30\ubc18\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
    2. \uac00\ub2a5\ub3c4(Likelihood)<\/strong>: \uc0ac\uac74 A\uac00 \ubc1c\uc0dd\ud55c \uacbd\uc6b0 \uc99d\uac70 B\uac00 \ubc1c\uc0dd\ud560 \ud655\ub960\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
    3. \uc8fc\ubcc0\ud655\ub960(Marginal Probability)<\/strong>: \uc99d\uac70 B\uac00 \ubc1c\uc0dd\ud560 \uc804\uccb4 \ud655\ub960\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4. \uc774\ub294 \ubaa8\ub4e0 \uac00\ub2a5\ud55c \uc0ac\uac74\uc744 \uace0\ub824\ud558\uc5ec B\uac00 \ubc1c\uc0dd\ud560 \ud655\ub960\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
    4. \uc0ac\ud6c4\ud655\ub960(Posterior Probability)<\/strong>: \uc0c8\ub85c\uc6b4 \uc99d\uac70 B\ub97c \uace0\ub824\ud558\uc5ec \uc0ac\uac74 A\uc758 \ud655\ub960\uc744 \uc5c5\ub370\uc774\ud2b8\ud569\ub2c8\ub2e4.<\/li>\n<\/ol>\n\n\n\n

      <\/span>\uc608\uc2dc<\/span><\/h3>\n\n\n\n

      \uc9c8\ubcd1 \uc9c4\ub2e8\uc5d0\uc11c \uc0ac\ud6c4\ud655\ub960\uc758 \uc0ac\uc6a9 \uc608\ub97c \ub4e4\uc5b4\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n

        \n
      • \uc0ac\uc804\ud655\ub960(Prior Probability)<\/strong>: \uc5b4\ub5a4 \uc9c8\ubcd1\uc774 \uc804\uccb4 \uc778\uad6c\uc5d0\uc11c \ubc1c\uc0dd\ud560 \ud655\ub960\uc774 1%\ub77c\uace0 \uac00\uc815\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
      • \uac00\ub2a5\ub3c4(Likelihood)<\/strong>: \uadf8 \uc9c8\ubcd1\uc5d0 \ub300\ud574 \ud2b9\uc815 \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ud655\ub960\uc774 90%\ub77c\uace0 \uac00\uc815\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
      • \uc8fc\ubcc0\ud655\ub960(Marginal Probability)<\/strong>: \uadf8 \uac80\uc0ac\uac00 \uc804\uccb4 \uc778\uad6c\uc5d0\uc11c \uc591\uc131\uc77c \ud655\ub960\uc774 10%\ub77c\uace0 \uac00\uc815\ud569\ub2c8\ub2e4.<\/li>\n<\/ul>\n\n\n\n

        \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \uc801\uc6a9\ud558\uc5ec, \uac80\uc0ac\uac00 \uc591\uc131\uc778 \uacbd\uc6b0 \uadf8 \uc0ac\ub78c\uc774 \uc2e4\uc81c\ub85c \uc9c8\ubcd1\uc5d0 \uac78\ub838\uc744 \ud655\ub960(\uc0ac\ud6c4\ud655\ub960)\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n

        \\[
        P(\\text{\uc9c8\ubcd1|\uc591\uc131}) = \\frac{P(\\text{\uc591\uc131|\uc9c8\ubcd1}) \\cdot P(\\text{\uc9c8\ubcd1})}{P(\\text{\uc591\uc131})}
        \\]<\/p>\n\n\n\n

        \uc989,<\/p>\n\n\n\n

        \\[
        P(\\text{\uc9c8\ubcd1|\uc591\uc131}) = \\frac{0.9 \\cdot 0.01}{0.1} = \\frac{0.009}{0.1} = 0.09
        \\]<\/p>\n\n\n\n

        \uac80\uc0ac\uac00 \uc591\uc131\uc778 \uacbd\uc6b0 \uadf8 \uc0ac\ub78c\uc774 \uc2e4\uc81c\ub85c \uc9c8\ubcd1\uc5d0 \uac78\ub838\uc744 \ud655\ub960\uc740 9%\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n

        <\/span>\uc0ac\ud6c4\ud655\ub960\uc758 \ud65c\uc6a9<\/span><\/h3>\n\n\n\n

        \uc0ac\ud6c4\ud655\ub960\uc740 \ub2e4\uc591\ud55c \ubd84\uc57c\uc5d0\uc11c \ud65c\uc6a9\ub429\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4:<\/p>\n\n\n\n

          \n
        • \uc758\ud559<\/strong>: \uc9c8\ubcd1 \uc9c4\ub2e8\uc5d0\uc11c \uac80\uc0ac\uc758 \uacb0\uacfc\ub97c \ud1a0\ub300\ub85c \uc9c8\ubcd1\uc758 \ud655\ub960\uc744 \uac31\uc2e0\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
        • \uae30\uacc4 \ud559\uc2b5<\/strong>: \ubaa8\ub378\uc758 \uc608\uce21\uc744 \uac1c\uc120\ud558\uae30 \uc704\ud574 \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\ub97c \uace0\ub824\ud569\ub2c8\ub2e4.<\/li>\n\n\n\n
        • \uacbd\uc81c\ud559<\/strong>: \uc0c8\ub85c\uc6b4 \uacbd\uc81c \ub370\uc774\ud130\ub97c \ubc18\uc601\ud558\uc5ec \uc608\uce21\uc744 \uc218\uc815\ud569\ub2c8\ub2e4.<\/li>\n<\/ul>\n\n\n\n

          <\/span>Python\uc744 \uc774\uc6a9\ud55c \uc0ac\ud6c4\ud655\ub960 \uacc4\uc0b0 \uc608\uc81c<\/span><\/h3>\n\n\n\n
          def bayes_theorem(prior_A, likelihood_B_given_A, marginal_B):\n    # \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc0ac\ud6c4\ud655\ub960 \uacc4\uc0b0\n    posterior_A_given_B = (likelihood_B_given_A * prior_A) \/ marginal_B\n    return posterior_A_given_B\n\n# \uc608\uc81c \uac12\nprior_A = 0.01  # \uc0ac\uc804\ud655\ub960: \uc9c8\ubcd1 \ubc1c\uc0dd \ud655\ub960\nlikelihood_B_given_A = 0.9  # \uac00\ub2a5\ub3c4: \uc9c8\ubcd1\uc774 \uc788\ub294 \uacbd\uc6b0 \uc591\uc131 \uac80\uc0ac \ud655\ub960\nmarginal_B = 0.1  # \uc8fc\ubcc0\ud655\ub960: \uc804\uccb4 \uc778\uad6c\uc5d0\uc11c \uc591\uc131 \uac80\uc0ac \ud655\ub960\n\n# \uc0ac\ud6c4\ud655\ub960 \uacc4\uc0b0\nposterior = bayes_theorem(prior_A, likelihood_B_given_A, marginal_B)\nprint(f\"\uc0ac\ud6c4\ud655\ub960: {posterior:.2f}\")<\/code><\/pre>\n\n\n\n
          \uc774 \ucf54\ub4dc\ub294 \uc0ac\uc804\ud655\ub960, \uac00\ub2a5\ub3c4, \uc8fc\ubcc0\ud655\ub960\uc744 \uc785\ub825\ubc1b\uc544 \uc0ac\ud6c4\ud655\ub960\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4. \uc704 \uc608\uc81c\uc5d0\uc11c\ub294 \uc0ac\ud6c4\ud655\ub960\uc774 0.09, \uc989 9%\ub85c \uacc4\uc0b0\ub429\ub2c8\ub2e4.<\/p>\n\n\n\n
          <\/span>\uc608\uc81c 1 : \ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8\uc744 \uc774\uc6a9\ud55c \uc0ac\ud6c4\ud655\ub960 \uacc4\uc0b0 \uc608\uc81c<\/span><\/h3>\n\n\n\n
          \"\"<\/figure>\n\n\n\n
          \uc704 \ubca4 \ub2e4\uc774\uc5b4\uadf8\ub7a8\uc740 \uc9c8\ubcd1\uacfc \uc591\uc131 \uac80\uc0ac \uacb0\uacfc\uc5d0 \ub300\ud55c \ud655\ub960\uc744 \uc2dc\uac01\uc801\uc73c\ub85c \ub098\ud0c0\ub0b8 \uac83\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n
            \n
          • \uc9c8\ubcd1(Disease, P(A))<\/strong>: \uc804\uccb4 \uc778\uad6c \uc911 \uc9c8\ubcd1\uc774 \uc788\ub294 \uc0ac\ub78c\uc758 \ube44\uc728 (1%).<\/li>\n\n\n\n
          • \uc591\uc131 \uac80\uc0ac(Positive Test, P(B))<\/strong>: \uc804\uccb4 \uc778\uad6c \uc911 \uc591\uc131 \uac80\uc0ac\ub97c \ubc1b\uc740 \uc0ac\ub78c\uc758 \ube44\uc728 (10%).<\/li>\n\n\n\n
          • \uc9c8\ubcd1\uacfc \uc591\uc131 \uac80\uc0ac(Disease and Positive Test, P(A and B))<\/strong>: \uc9c8\ubcd1\uc774 \uc788\uace0 \uc591\uc131 \uac80\uc0ac\ub97c \ubc1b\uc740 \uc0ac\ub78c\uc758 \ube44\uc728 (\uc9c8\ubcd1\uc774 \uc788\ub294 \uacbd\uc6b0 \uc591\uc131 \uac80\uc0ac\ub97c \ubc1b\uc744 \ud655\ub960 90% * \uc9c8\ubcd1\uc774 \uc788\uc744 \ud655\ub960 1%).<\/li>\n<\/ul>\n\n\n\n
            \uc774 \ub2e4\uc774\uc5b4\uadf8\ub7a8\uc744 \ud1b5\ud574 \uc0c8\ub85c\uc6b4 \uc815\ubcf4(\uc591\uc131 \uac80\uc0ac \uacb0\uacfc)\ub97c \uace0\ub824\ud558\uc5ec \uc0ac\ud6c4\ud655\ub960\uc744 \uacc4\uc0b0\ud558\ub294 \uacfc\uc815\uc744 \uc774\ud574\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
            <\/span>\uc608\uc81c 2 : \uc790\ub3d9\ucc28 \uacf5\uc7a5\uc5d0\uc11c\uc758 \uc0ac\ud6c4\ud654\ub960<\/span><\/h3>\n\n\n\n
            <\/span>\uc0ac\ud6c4\ud655\ub960\uacfc \ubca0\uc774\uc988 \uc815\ub9ac\uc758 \ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8 \uc124\uba85<\/span><\/h3>\n\n\n\n
            \uc790\ub3d9\ucc28 \uacf5\uc7a5\uc5d0\uc11c \ubd88\ub7c9 \uc81c\ud488\uc758 \uc0ac\ud6c4\ud655\ub960\uc744 \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \ud1b5\ud574 \uc124\uba85\ud558\uace0, \uc774\ub97c \ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8\uc73c\ub85c \uc2dc\uac01\ud654\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
            \uc608\uc2dc \uc0c1\ud669<\/h4>\n\n\n\n
              \n
            1. \uc0ac\uc804\ud655\ub960 (P(\ubd88\ub7c9))<\/strong>: \uacf5\uc7a5\uc5d0\uc11c \uc0dd\uc0b0\ub41c \uc81c\ud488 \uc911 \ubd88\ub7c9\uc77c \ud655\ub960 = 2% (0.02).<\/li>\n\n\n\n
            2. \uac80\uc0ac\uc758 \ubbfc\uac10\ub3c4 (P(\uc591\uc131 | \ubd88\ub7c9))<\/strong>: \ubd88\ub7c9 \uc81c\ud488\uc774 \uc788\uc744 \ub54c \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ud655\ub960 = 95% (0.95).<\/li>\n\n\n\n
            3. \uac80\uc0ac\uc758 \uc704\uc591\uc131\uc728 (P(\uc591\uc131 | \uc815\uc0c1))<\/strong>: \uc815\uc0c1 \uc81c\ud488\uc774\uc9c0\ub9cc \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ud655\ub960 = 10% (0.10).<\/li>\n\n\n\n
            4. \uc591\uc131\ub960 (P(\uc591\uc131))<\/strong>: \uc804\uccb4 \uc81c\ud488\uc5d0\uc11c \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc77c \ud655\ub960.<\/li>\n<\/ol>\n\n\n\n
              \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \ud1b5\ud574 \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \ubd88\ub7c9\uc77c \ud655\ub960(P(\ubd88\ub7c9 | \uc591\uc131))\uc744 \uacc4\uc0b0\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
              \ubca0\uc774\uc988 \uc815\ub9ac \uc801\uc6a9<\/h4>\n\n\n\n
              \\[
              P(\\text{\ubd88\ub7c9|\uc591\uc131}) = \\frac{P(\\text{\uc591\uc131|\ubd88\ub7c9}) \\cdot P(\\text{\ubd88\ub7c9})}{P(\\text{\uc591\uc131})}
              \\]<\/p>\n\n\n\n
              \uc5ec\uae30\uc11c P(\uc591\uc131)\ub294 \uc804\uccb4 \uc81c\ud488\uc5d0\uc11c \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc77c \ud655\ub960\ub85c, \ub2e4\uc74c\uacfc \uac19\uc774 \uacc4\uc0b0\ub429\ub2c8\ub2e4:<\/p>\n\n\n\n
              \\[
              P(\\text{\uc591\uc131}) = P(\\text{\uc591\uc131|\ubd88\ub7c9}) \\cdot P(\\text{\ubd88\ub7c9}) + P(\\text{\uc591\uc131|\uc815\uc0c1}) \\cdot P(\\text{\uc815\uc0c1})
              \\]<\/p>\n\n\n\n
              P(\uc815\uc0c1)\ub294 1 – P(\ubd88\ub7c9)\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n
              \\[
              P(\\text{\uc591\uc131}) = (0.95 \\cdot 0.02) + (0.10 \\cdot 0.98) = 0.019 + 0.098 = 0.117
              \\]<\/p>\n\n\n\n
              \ub530\ub77c\uc11c, \uc0ac\ud6c4\ud655\ub960 P(\ubd88\ub7c9|\uc591\uc131)\uc740 \ub2e4\uc74c\uacfc \uac19\uc774 \uacc4\uc0b0\ub429\ub2c8\ub2e4:<\/p>\n\n\n\n
              \\[
              P(\\text{\ubd88\ub7c9|\uc591\uc131}) = \\frac{0.95 \\cdot 0.02}{0.117} \\approx 0.162
              \\]<\/p>\n\n\n\n
              \uc989, \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \ubd88\ub7c9\uc77c \ud655\ub960\uc740 \uc57d 16.2%\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n
              <\/span>\ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8\uc744 \ud1b5\ud55c \uc2dc\uac01\ud654<\/span><\/h3>\n\n\n\n
                \n
              1. \uc804\uccb4 \uc81c\ud488 \uc9d1\ud569<\/strong>: \uc804\uccb4 \uc81c\ud488 \uc911 \ubd88\ub7c9 \uc81c\ud488\uacfc \uc815\uc0c1 \uc81c\ud488.<\/li>\n\n\n\n
              2. \uc591\uc131 \uac80\uc0ac \uc9d1\ud569<\/strong>: \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc778 \uc81c\ud488.<\/li>\n<\/ol>\n\n\n\n
                \uc774 \ub450 \uc9d1\ud569\uc774 \uacb9\uce58\ub294 \ubd80\ubd84\uc744 \ud1b5\ud574 \uc0ac\ud6c4\ud655\ub960\uc744 \uc2dc\uac01\ud654\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
                \ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8\uc744 \uadf8\ub824\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
                import matplotlib.pyplot as plt\nfrom matplotlib_venn import venn2\n\n# Define the subsets\ntotal_products = 1.0\ndefective_products = 0.02\nnon_defective_products = total_products - defective_products\npositive_test_defective = 0.95 * defective_products\npositive_test_non_defective = 0.10 * non_defective_products\npositive_test_total = positive_test_defective + positive_test_non_defective\n\n# Sizes for Venn diagram\nsizes = {\n    '10': positive_test_non_defective,  # Positive but not defective\n    '01': positive_test_defective,      # Defective but not positive\n    '11': positive_test_defective       # Both positive and defective\n}\n\n# Create Venn diagram\nplt.figure(figsize=(10, 7))\nvenn = venn2(subsets=sizes, set_labels=('Positive Test', 'Defective Product'))\nvenn.get_label_by_id('10').set_text('False Positive')\nvenn.get_label_by_id('01').set_text('False Negative')\nvenn.get_label_by_id('11').set_text('True Positive')\n\nplt.title(\"Venn Diagram of Positive Test and Defective Products\")\nplt.show()<\/code><\/pre>\n\n\n\n
                \uc774 \uadf8\ub9bc\uc740 \uc804\uccb4 \uc81c\ud488 \uc911\uc5d0\uc11c \uc591\uc131 \uac80\uc0ac \uacb0\uacfc\ub97c \ubc1b\uc740 \uc81c\ud488\uacfc \uc2e4\uc81c\ub85c \ubd88\ub7c9\uc778 \uc81c\ud488\uc758 \uad00\uacc4\ub97c \uc2dc\uac01\uc801\uc73c\ub85c \ub098\ud0c0\ub0c5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \uac80\uc0ac \uacb0\uacfc\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \ubd88\ub7c9\uc77c \ud655\ub960\uc744 \uc2dc\uac01\uc801\uc73c\ub85c \uc774\ud574\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
                <\/span>\uc608\uc81c 3 : \ubca4\ub2e4\uc774\uc5b4\uadf8\ub7a8\uc744 \ud1b5\ud55c \uc2dc\uac01\ud654<\/span><\/h3>\n\n\n\n
                \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc0ac\ud6c4 \ud655\ub960\uc744 \uacc4\uc0b0\ud558\ub294 \uc608\uc81c\ub97c \ubcf4\uc5ec\ub4dc\ub9ac\uaca0\uc2b5\ub2c8\ub2e4. \uc608\uc81c\ub294 \uc758\ub8cc \uc9c4\ub2e8 \uc2dc\ub098\ub9ac\uc624\ub97c \ubc14\ud0d5\uc73c\ub85c \ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
                <\/span>\uc2dc\ub098\ub9ac\uc624<\/span><\/h3>\n\n\n\n
                \uc5b4\ub5a4 \uc9c8\ubcd1\uc5d0 \ub300\ud574 \uac80\uc0ac\ud558\ub294 \uacbd\uc6b0\ub97c \uc0dd\uac01\ud574 \ubd05\uc2dc\ub2e4.<\/p>\n\n\n\n
                  \n
                • \ud2b9\uc815 \uc9c8\ubcd1\uc5d0 \uac78\ub9b4 \uc0ac\uc804 \ud655\ub960 \\(( P(A) )\\)\uc740 1% (0.01)\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
                • \uc9c8\ubcd1\uc774 \uc788\ub294 \uacbd\uc6b0 \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ud655\ub960 \\(( P(B|A) )\\)\uc740 99% (0.99)\uc785\ub2c8\ub2e4.<\/li>\n\n\n\n
                • \uc9c8\ubcd1\uc774 \uc5c6\ub294 \uacbd\uc6b0 \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ud655\ub960 \\(( P(B|~A) )\\)\uc740 5% (0.05)\uc785\ub2c8\ub2e4.<\/li>\n<\/ul>\n\n\n\n
                  \uc774\uc81c, \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \uc9c8\ubcd1\uc774 \uc788\uc744 \ud655\ub960 (( P(A|B) ))\uc744 \uacc4\uc0b0\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
                  <\/span>\ubca0\uc774\uc988 \uc815\ub9ac \uacf5\uc2dd<\/span><\/h3>\n\n\n\n
                  \ubca0\uc774\uc988 \uc815\ub9ac\ub294 \ub2e4\uc74c\uacfc \uac19\uc774 \uc815\uc758\ub429\ub2c8\ub2e4:<\/p>\n\n\n\n
                  \\[
                  P(A|B) = \\frac{P(B|A) \\cdot P(A)}{P(B)}
                  \\]<\/p>\n\n\n\n
                  \uc5ec\uae30\uc11c ( P(B) )\ub294 \uc804\uccb4 \uc591\uc131 \uac80\uc0ac \ud655\ub960\ub85c, \ub2e4\uc74c\uacfc \uac19\uc774 \uacc4\uc0b0\ub429\ub2c8\ub2e4:<\/p>\n\n\n\n
                  \\[
                  P(B) = P(B|A) \\cdot P(A) + P(B|~A) \\cdot P(~A)
                  \\]<\/p>\n\n\n\n
                  <\/span>\uacc4\uc0b0 \uacfc\uc815<\/span><\/h3>\n\n\n\n
                    \n
                  1. \uc0ac\uc804 \ud655\ub960 \\(( P(A) )\\)<\/strong>: 0.01<\/li>\n\n\n\n
                  2. \uc9c8\ubcd1\uc774 \uc788\uc744 \ub54c \uc591\uc131\uc77c \ud655\ub960 \\(( P(B|A) )\\)<\/strong>: 0.99<\/li>\n\n\n\n
                  3. \uc9c8\ubcd1\uc774 \uc5c6\uc744 \ub54c \uc591\uc131\uc77c \ud655\ub960 \\(( P(B|~A) )\\)<\/strong>: 0.05<\/li>\n\n\n\n
                  4. \uc9c8\ubcd1\uc774 \uc5c6\uc744 \ud655\ub960 \\(( P(~A) ))<\/strong>: 1 – ( P(A) \\) = 0.99<\/li>\n<\/ol>\n\n\n\n
                    \\[
                    P(B) = (0.99 \\times 0.01) + (0.05 \\times 0.99)
                    \\]<\/p>\n\n\n\n
                    \\[
                    P(B) = 0.0099 + 0.0495 = 0.0594
                    \\]<\/p>\n\n\n\n
                    \uc774\uc81c \ubca0\uc774\uc988 \uc815\ub9ac\ub97c \uc801\uc6a9\ud558\uc5ec ( P(A|B) )\ub97c \uacc4\uc0b0\ud569\ub2c8\ub2e4:<\/p>\n\n\n\n
                    \\[
                    P(A|B) = \\frac{0.99 \\times 0.01}{0.0594} = \\frac{0.0099}{0.0594} \\approx 0.1667
                    \\]<\/p>\n\n\n\n
                    \ub530\ub77c\uc11c, \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \uc9c8\ubcd1\uc774 \uc788\uc744 \ud655\ub960\uc740 \uc57d 16.67%\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n
                    <\/span>Python \ucf54\ub4dc \uc608\uc81c<\/span><\/h3>\n\n\n\n
                    \uc704 \uacc4\uc0b0\uc744 Python \ucf54\ub4dc\ub85c \uc791\uc131\ud574 \ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
                    # Given probabilities\nP_A = 0.01  # Prior probability of having the disease\nP_B_given_A = 0.99  # Probability of testing positive given the disease\nP_B_given_not_A = 0.05  # Probability of testing positive given no disease\n\n# Complementary probability\nP_not_A = 1 - P_A\n\n# Total probability of testing positive\nP_B = P_B_given_A * P_A + P_B_given_not_A * P_not_A\n\n# Posterior probability\nP_A_given_B = (P_B_given_A * P_A) \/ P_B\n\nprint(f\"Total probability of testing positive, P(B): {P_B:.4f}\")\nprint(f\"Posterior probability of having the disease given a positive test, P(A|B): {P_A_given_B:.4f}\")<\/code><\/pre>\n\n\n\n
                    \uc774 \ucf54\ub4dc\ub97c \uc2e4\ud589\ud558\uba74 \ub2e4\uc74c\uacfc \uac19\uc740 \uacb0\uacfc\uac00 \ub098\uc635\ub2c8\ub2e4:<\/p>\n\n\n\n
                    Total probability of testing positive, P(B): 0.0594\nPosterior probability of having the disease given a positive test, P(A|B): 0.1667<\/code><\/pre>\n\n\n\n
                    \uc774\ub97c \ud1b5\ud574, \uac80\uc0ac\uac00 \uc591\uc131\uc77c \ub54c \uc2e4\uc81c\ub85c \uc9c8\ubcd1\uc774 \uc788\uc744 \ud655\ub960\uc744 \uad6c\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n","protected":false},"excerpt":{"rendered":"
                    \uc0ac\ud6c4\ud655\ub960(Posterior Probability)\uc740 \ubca0\uc774\uc988 \uc815\ub9ac(Bayes’ theorem)\uc5d0 \uae30\ubc18\ud55c \uac1c\ub150\uc73c\ub85c, \ud2b9\uc815 \uc0ac\uac74\uc774 \ubc1c\uc0dd\ud55c \ud6c4\uc5d0 \uadf8 \uc0ac\uac74\uc5d0 \ub300\ud55c \uc870\uac74\ubd80 \ud655\ub960\uc744 \uc5c5\ub370\uc774\ud2b8\ud558\ub294 \uacfc\uc815\uc785\ub2c8\ub2e4….<\/p>\n","protected":false},"author":1,"featured_media":983,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[],"class_list":["post-774","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-stat","wpcat-6-id"],"_links":{"self":[{"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/posts\/774","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=774"}],"version-history":[{"count":4,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/posts\/774\/revisions"}],"predecessor-version":[{"id":989,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/posts\/774\/revisions\/989"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=\/wp\/v2\/media\/983"}],"wp:attachment":[{"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=774"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=774"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.inmining.co.kr\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=774"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}