\n \uae08\uc735 \uc2dc\uc7a5\uc5d0\uc11c\uc758 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18 \uc758\uc0ac\uacb0\uc815\uc744 \uac00\ub2a5\ud558\uac8c \ud558\uc5ec \ud22c\uc790\uc790\ub4e4\uc5d0\uac8c \ud070 \uc774\uc810\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4.\n \ud2b9\ud788 \uba38\uc2e0\ub7ec\ub2dd(Machine Learning)\uacfc \ub525\ub7ec\ub2dd(Deep Learning) \uae30\uc220\uc758 \ubc1c\uc804\uc740 \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc758 \uc124\uacc4 \ubc0f \uac1c\uc120\uc5d0 \ud601\uc2e0\uc801\uc778 \ubcc0\ud654\ub97c \uac00\uc838\uc654\uc2b5\ub2c8\ub2e4.\n \ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294 LightGBM<\/strong>\uc744 \ud65c\uc6a9\ud558\uc5ec \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc18\uc758 \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc744 \ub9cc\ub4dc\ub294 \ubc29\ubc95\uc744 \ub2e4\ub8e8\uaca0\uc2b5\ub2c8\ub2e4. LightGBM\uc740 Gradient Boosting Decision Tree (GBDT) \uc54c\uace0\ub9ac\uc998\uc758 \ubcc0\uc885\uc73c\ub85c,\n \ub300\ub7c9\uc758 \ub370\uc774\ud130 \ucc98\ub9ac\uc640 \ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4\uac00 \uc7a5\uc810\uc785\ub2c8\ub2e4.\n <\/p>\n2. \uba38\uc2e0\ub7ec\ub2dd\uc758 \uac1c\uc694<\/h2>\n\n \uba38\uc2e0\ub7ec\ub2dd\uc740 \ub370\uc774\ud130\ub97c \ud1b5\ud574 \uc790\ub3d9\uc73c\ub85c \ud328\ud134\uc744 \ud559\uc2b5\ud558\uace0 \uc608\uce21\uc744 \uc218\ud589\ud558\ub294 \uae30\uc220\uc785\ub2c8\ub2e4.\n \uae08\uc735 \ubd84\uc57c\uc5d0\uc11c\ub294 \uc2dc\uacc4\uc5f4 \ub370\uc774\ud130, \uc9c0\ud45c, \ub274\uc2a4 \ub4f1 \ub2e4\uc591\ud55c \ud615\ud0dc\uc758 \ub370\uc774\ud130\ub97c \uae30\ubc18\uc73c\ub85c \uc8fc\uc2dd \uac00\uaca9 \uc608\uce21, \uc704\ud5d8 \uad00\ub9ac, \uc804\ub7b5 \ucd5c\uc801\ud654 \ub4f1\uc758 \ubb38\uc81c\ub97c \ud574\uacb0\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n\n\uac10\ub3c5 \ud559\uc2b5(Supervised Learning): \uc8fc\uc5b4\uc9c4 \uc785\ub825 \ub370\uc774\ud130\uc5d0 \ub300\ud55c \uc815\ub2f5(\ucd9c\ub825)\uc744 \ud559\uc2b5\ud558\ub294 \ubc29\uc2dd\uc785\ub2c8\ub2e4.<\/li>\n\ube44\uac10\ub3c5 \ud559\uc2b5(Unsupervised Learning): \ub808\uc774\ube14\uc774 \uc5c6\ub294 \ub370\uc774\ud130\uc5d0\uc11c \ud328\ud134\uc744 \ucc3e\ub294 \ud559\uc2b5 \ubc29\uc2dd\uc785\ub2c8\ub2e4.<\/li>\n\uac15\ud654 \ud559\uc2b5(Reinforcement Learning): \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uacfc \uc0c1\ud638\uc791\uc6a9\ud558\uba70 \ubcf4\uc0c1\uc744 \ucd5c\ub300\ud654\ud558\ub294 \ubc29\ud5a5\uc73c\ub85c \ud559\uc2b5\ud558\ub294 \ubc29\uc2dd\uc785\ub2c8\ub2e4.<\/li>\n<\/ul>\n3. LightGBM \uc18c\uac1c<\/h2>\n\n LightGBM\uc740 Microsoft\uc5d0\uc11c \uac1c\ubc1c\ud55c Gradient Boosting Framework\uc785\ub2c8\ub2e4.\n \ud2b9\ud788 \ub300\uaddc\ubaa8 \ub370\uc774\ud130\uc14b\uc5d0 \uc801\ud569\ud558\uba70, \uba38\uc2e0\ub7ec\ub2dd \ub300\ud68c<\/strong>\uc640 \uc2e4\uc81c \uc0b0\uc5c5<\/strong>\uc5d0\uc11c \ub110\ub9ac \uc0ac\uc6a9\ub429\ub2c8\ub2e4.\n LightGBM\uc758 \uc8fc\uc694 \ud2b9\uc9d5 \uc911 \ud558\ub098\ub294 \ub9ac\ud504 \uc911\uc2ec\uc758 \ud2b8\ub9ac \uc131\uc7a5(Leaf-wise tree growth)<\/strong> \ubc29\uc2dd\uc785\ub2c8\ub2e4.\n \uc774\ub294 \ubaa8\ub378\uc758 \uc815\ud655\uc131\uc744 \ub192\uc774\uba74\uc11c\ub3c4 \uacc4\uc0b0 \uc18d\ub3c4\ub97c \uc99d\uac00\uc2dc\ud0b5\ub2c8\ub2e4.\n <\/p>\n3.1 LightGBM\uc758 \uc7a5\uc810<\/h3>\n\n\ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4: \ub300\ub7c9\uc758 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\uba74\uc11c\ub3c4 \ube60\ub974\uac8c \ud559\uc2b5 \uac00\ub2a5\ud569\ub2c8\ub2e4.<\/li>\n\uba54\ubaa8\ub9ac \ud6a8\uc728\uc131: \uba54\ubaa8\ub9ac\ub97c \ud6a8\uc728\uc801\uc73c\ub85c \uc0ac\uc6a9\ud558\uc5ec \ud070 \ub370\uc774\ud130\uc14b\uc744 \ub2e4\ub8f0 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/li>\n\ub192\uc740 \uc815\ud655\ub3c4: GBDT\uc758 \uc7a5\uc810\uc744 \uadf9\ub300\ud654\ud558\uc5ec \ub192\uc740 \uc608\uce21 \uc131\ub2a5\uc744 \uc790\ub791\ud569\ub2c8\ub2e4.<\/li>\n<\/ul>\n4. ML4T(Machine Learning for Trading)\ub780?<\/h2>\n\n ML4T\ub294 \uba38\uc2e0\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5 \uc218\ub9bd \ubc0f \ucd5c\uc801\ud654 \uc791\uc5c5\uc744 \uc758\ubbf8\ud569\ub2c8\ub2e4.\n \uc0ac\uc6a9\uc790\ub294 \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc95\uc744 \ud1b5\ud574 \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uace0, \uc774\ub97c \ud1b5\ud574 \ubcf4\ub2e4 \ud6a8\uacfc\uc801\uc778 \uc758\uc0ac\uacb0\uc815\uc744 \ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5. LightGBM\uc744 \ud65c\uc6a9\ud55c \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uad6c\ucd95<\/h2>\n5.1 \ub370\uc774\ud130 \uc218\uc9d1<\/h3>\n\n \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uae30 \uc704\ud574\uc11c\ub294 \uc6b0\uc120\uc801\uc73c\ub85c \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4.\n \uc8fc\uac00 \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud558\uae30 \uc704\ud574\uc11c\ub294 API\ub97c \uc0ac\uc6a9\ud558\uac70\ub098 \uae08\uc735 \ub370\uc774\ud130 \uc81c\uacf5 \uc5c5\uccb4\uc758 \ub3c4\uc6c0\uc744 \ubc1b\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5.2 \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h3>\n\n \uc218\uc9d1\ud55c \ub370\uc774\ud130\ub294 \ubaa8\ub378 \ud559\uc2b5\uc5d0 \uc801\ud569\ud55c \ud615\uc2dd\uc73c\ub85c \ubcc0\ud658\ub418\uc5b4\uc57c \ud569\ub2c8\ub2e4.\n \uc774 \uacfc\uc815\uc5d0\uc11c \uacb0\uce21\uce58\ub97c \ucc98\ub9ac\ud558\uace0, \ud2b9\uc131 \uc5d4\uc9c0\ub2c8\uc5b4\ub9c1\uc744 \ud1b5\ud574 \uae30\uc874 \ub370\uc774\ud130\ub85c\ubd80\ud130 \uc0c8\ub85c\uc6b4 \ud2b9\uc9d5\uc744 \uc0dd\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5.3 \ubaa8\ub378 \ud559\uc2b5<\/h3>\n\n \uc804\ucc98\ub9ac\ub41c \ub370\uc774\ud130\ub97c \uae30\ubc18\uc73c\ub85c LightGBM \ubaa8\ub378\uc744 \ud559\uc2b5\ud569\ub2c8\ub2e4.\n \uc544\ub798\ub294 Python\uc744 \uc0ac\uc6a9\ud55c LightGBM \ubaa8\ub378\uc758 \uae30\ubcf8\uc801\uc778 \ud559\uc2b5 \ucf54\ub4dc\uc785\ub2c8\ub2e4:\n <\/p>\n\nimport lightgbm as lgb\nfrom sklearn.model_selection import train_test_split\n\n# \ub370\uc774\ud130\uc14b \ub85c\ub4dc\ndata = ... # \ub370\uc774\ud130 \ub85c\ub4dc \ucf54\ub4dc\nX = data.drop(columns='target')\ny = data['target']\n\n# \ud6c8\ub828 \ub370\uc774\ud130\uc640 \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub85c \ubd84\ud560\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\n# LightGBM \ub370\uc774\ud130\uc14b \uc0dd\uc131\ntrain_data = lgb.Dataset(X_train, label=y_train)\ntest_data = lgb.Dataset(X_test, label=y_test)\n\n# \ubaa8\ub378 \ud30c\ub77c\ubbf8\ud130 \uc124\uc815\nparams = {\n 'objective': 'binary',\n 'metric': 'auc',\n 'learning_rate': 0.05,\n 'num_leaves': 31,\n 'verbose': -1\n}\n\n# \ubaa8\ub378 \ud559\uc2b5\nmodel = lgb.train(params, train_data, num_boost_round=1000, valid_sets=[test_data], early_stopping_rounds=100)\n <\/code><\/pre>\n5.4 \ubaa8\ub378 \ud3c9\uac00<\/h3>\n\n \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uae30 \uc704\ud574 AUC(Area Under the Curve) \uc9c0\ud45c\ub97c \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n \ud3c9\uac00\ub41c \uc131\ub2a5\uc744 \uae30\ubc18\uc73c\ub85c \ubaa8\ub378\uc758 \ud30c\ub77c\ubbf8\ud130\ub97c \uc870\uc815\ud558\uace0, \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc744 \ud1b5\ud574 \ucd5c\uc801\uc758 \uc131\ub2a5\uc744 \ucc3e\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4.\n <\/p>\n5.5 \uc804\ub7b5 \uc2e4\ud589<\/h3>\n\n \ucd5c\uc801\uc758 \ubaa8\ub378\uc744 \ubc14\ud0d5\uc73c\ub85c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc744 \uc2e4\ud589\ud569\ub2c8\ub2e4.\n \uc774 \ub2e8\uacc4\uc5d0\uc11c\ub294 \uc2e4\uc2dc\uac04 \ub370\uc774\ud130 \uc2a4\ud2b8\ub9ac\ubc0d\uc744 \ubc1b\uc544 \ubaa8\ub378\uc744 \ud1b5\ud574 \ub9e4\uc218\/\ub9e4\ub3c4 \uc2e0\ud638\ub97c \ubc1c\uc0dd\uc2dc\ud0a4\uace0, \uc774\ub97c \uae30\ubc18\uc73c\ub85c \uc790\ub3d9\uc73c\ub85c \uac70\ub798\ub97c \uc218\ud589\ud558\ub294 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n6. \uacb0\ub860<\/h2>\n\n \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18 \uacb0\uc815 \uacfc\uc815\uc744 \ud1b5\ud574 \uc804\ud1b5\uc801\uc778 \ud2b8\ub808\uc774\ub529 \ubc29\uc2dd\uc5d0 \ube44\ud574\n \ub9ce\uc740 \uc774\uc810\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4. \ud2b9\ud788 LightGBM\uc740 \ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4\uc640 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc81c\uacf5\ud558\uc5ec,\n \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uac1c\ubc1c\uc5d0 \uc720\uc6a9\ud55c \ub3c4\uad6c\uc785\ub2c8\ub2e4. \uc9c0\uc18d\uc801\uc73c\ub85c \uc54c\uace0\ub9ac\uc998\uc744 \uac1c\uc120\ud558\uace0, \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\uc640 \uc804\ub7b5\uc744 \uc801\uc6a9\ud558\uc5ec\n \uc548\uc815\uc801\uc774\uace0 \uc218\uc775\uc131 \uc788\ub294 \ud2b8\ub808\uc774\ub529\uc744 \uc2e4\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n7. \ucc38\uace0 \uc790\ub8cc<\/h2>\n\nLightGBM Documentation<\/a><\/li>\nCoursera – Machine Learning by Andrew Ng<\/a><\/li>\nTowards Data Science<\/a> (\ub370\uc774\ud130 \uc0ac\uc774\uc5b8\uc2a4 \uad00\ub828 \ube14\ub85c\uadf8)<\/li>\n<\/ul>\n
\n \uba38\uc2e0\ub7ec\ub2dd\uc740 \ub370\uc774\ud130\ub97c \ud1b5\ud574 \uc790\ub3d9\uc73c\ub85c \ud328\ud134\uc744 \ud559\uc2b5\ud558\uace0 \uc608\uce21\uc744 \uc218\ud589\ud558\ub294 \uae30\uc220\uc785\ub2c8\ub2e4.\n \uae08\uc735 \ubd84\uc57c\uc5d0\uc11c\ub294 \uc2dc\uacc4\uc5f4 \ub370\uc774\ud130, \uc9c0\ud45c, \ub274\uc2a4 \ub4f1 \ub2e4\uc591\ud55c \ud615\ud0dc\uc758 \ub370\uc774\ud130\ub97c \uae30\ubc18\uc73c\ub85c \uc8fc\uc2dd \uac00\uaca9 \uc608\uce21, \uc704\ud5d8 \uad00\ub9ac, \uc804\ub7b5 \ucd5c\uc801\ud654 \ub4f1\uc758 \ubb38\uc81c\ub97c \ud574\uacb0\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n LightGBM\uc740 Microsoft\uc5d0\uc11c \uac1c\ubc1c\ud55c Gradient Boosting Framework\uc785\ub2c8\ub2e4.\n \ud2b9\ud788 \ub300\uaddc\ubaa8 \ub370\uc774\ud130\uc14b\uc5d0 \uc801\ud569\ud558\uba70, \uba38\uc2e0\ub7ec\ub2dd \ub300\ud68c<\/strong>\uc640 \uc2e4\uc81c \uc0b0\uc5c5<\/strong>\uc5d0\uc11c \ub110\ub9ac \uc0ac\uc6a9\ub429\ub2c8\ub2e4.\n LightGBM\uc758 \uc8fc\uc694 \ud2b9\uc9d5 \uc911 \ud558\ub098\ub294 \ub9ac\ud504 \uc911\uc2ec\uc758 \ud2b8\ub9ac \uc131\uc7a5(Leaf-wise tree growth)<\/strong> \ubc29\uc2dd\uc785\ub2c8\ub2e4.\n \uc774\ub294 \ubaa8\ub378\uc758 \uc815\ud655\uc131\uc744 \ub192\uc774\uba74\uc11c\ub3c4 \uacc4\uc0b0 \uc18d\ub3c4\ub97c \uc99d\uac00\uc2dc\ud0b5\ub2c8\ub2e4.\n <\/p>\n3.1 LightGBM\uc758 \uc7a5\uc810<\/h3>\n\n\ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4: \ub300\ub7c9\uc758 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\uba74\uc11c\ub3c4 \ube60\ub974\uac8c \ud559\uc2b5 \uac00\ub2a5\ud569\ub2c8\ub2e4.<\/li>\n\uba54\ubaa8\ub9ac \ud6a8\uc728\uc131: \uba54\ubaa8\ub9ac\ub97c \ud6a8\uc728\uc801\uc73c\ub85c \uc0ac\uc6a9\ud558\uc5ec \ud070 \ub370\uc774\ud130\uc14b\uc744 \ub2e4\ub8f0 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/li>\n\ub192\uc740 \uc815\ud655\ub3c4: GBDT\uc758 \uc7a5\uc810\uc744 \uadf9\ub300\ud654\ud558\uc5ec \ub192\uc740 \uc608\uce21 \uc131\ub2a5\uc744 \uc790\ub791\ud569\ub2c8\ub2e4.<\/li>\n<\/ul>\n4. ML4T(Machine Learning for Trading)\ub780?<\/h2>\n\n ML4T\ub294 \uba38\uc2e0\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5 \uc218\ub9bd \ubc0f \ucd5c\uc801\ud654 \uc791\uc5c5\uc744 \uc758\ubbf8\ud569\ub2c8\ub2e4.\n \uc0ac\uc6a9\uc790\ub294 \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc95\uc744 \ud1b5\ud574 \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uace0, \uc774\ub97c \ud1b5\ud574 \ubcf4\ub2e4 \ud6a8\uacfc\uc801\uc778 \uc758\uc0ac\uacb0\uc815\uc744 \ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5. LightGBM\uc744 \ud65c\uc6a9\ud55c \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uad6c\ucd95<\/h2>\n5.1 \ub370\uc774\ud130 \uc218\uc9d1<\/h3>\n\n \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uae30 \uc704\ud574\uc11c\ub294 \uc6b0\uc120\uc801\uc73c\ub85c \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4.\n \uc8fc\uac00 \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud558\uae30 \uc704\ud574\uc11c\ub294 API\ub97c \uc0ac\uc6a9\ud558\uac70\ub098 \uae08\uc735 \ub370\uc774\ud130 \uc81c\uacf5 \uc5c5\uccb4\uc758 \ub3c4\uc6c0\uc744 \ubc1b\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5.2 \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h3>\n\n \uc218\uc9d1\ud55c \ub370\uc774\ud130\ub294 \ubaa8\ub378 \ud559\uc2b5\uc5d0 \uc801\ud569\ud55c \ud615\uc2dd\uc73c\ub85c \ubcc0\ud658\ub418\uc5b4\uc57c \ud569\ub2c8\ub2e4.\n \uc774 \uacfc\uc815\uc5d0\uc11c \uacb0\uce21\uce58\ub97c \ucc98\ub9ac\ud558\uace0, \ud2b9\uc131 \uc5d4\uc9c0\ub2c8\uc5b4\ub9c1\uc744 \ud1b5\ud574 \uae30\uc874 \ub370\uc774\ud130\ub85c\ubd80\ud130 \uc0c8\ub85c\uc6b4 \ud2b9\uc9d5\uc744 \uc0dd\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n5.3 \ubaa8\ub378 \ud559\uc2b5<\/h3>\n\n \uc804\ucc98\ub9ac\ub41c \ub370\uc774\ud130\ub97c \uae30\ubc18\uc73c\ub85c LightGBM \ubaa8\ub378\uc744 \ud559\uc2b5\ud569\ub2c8\ub2e4.\n \uc544\ub798\ub294 Python\uc744 \uc0ac\uc6a9\ud55c LightGBM \ubaa8\ub378\uc758 \uae30\ubcf8\uc801\uc778 \ud559\uc2b5 \ucf54\ub4dc\uc785\ub2c8\ub2e4:\n <\/p>\n\nimport lightgbm as lgb\nfrom sklearn.model_selection import train_test_split\n\n# \ub370\uc774\ud130\uc14b \ub85c\ub4dc\ndata = ... # \ub370\uc774\ud130 \ub85c\ub4dc \ucf54\ub4dc\nX = data.drop(columns='target')\ny = data['target']\n\n# \ud6c8\ub828 \ub370\uc774\ud130\uc640 \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub85c \ubd84\ud560\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\n# LightGBM \ub370\uc774\ud130\uc14b \uc0dd\uc131\ntrain_data = lgb.Dataset(X_train, label=y_train)\ntest_data = lgb.Dataset(X_test, label=y_test)\n\n# \ubaa8\ub378 \ud30c\ub77c\ubbf8\ud130 \uc124\uc815\nparams = {\n 'objective': 'binary',\n 'metric': 'auc',\n 'learning_rate': 0.05,\n 'num_leaves': 31,\n 'verbose': -1\n}\n\n# \ubaa8\ub378 \ud559\uc2b5\nmodel = lgb.train(params, train_data, num_boost_round=1000, valid_sets=[test_data], early_stopping_rounds=100)\n <\/code><\/pre>\n5.4 \ubaa8\ub378 \ud3c9\uac00<\/h3>\n\n \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uae30 \uc704\ud574 AUC(Area Under the Curve) \uc9c0\ud45c\ub97c \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n \ud3c9\uac00\ub41c \uc131\ub2a5\uc744 \uae30\ubc18\uc73c\ub85c \ubaa8\ub378\uc758 \ud30c\ub77c\ubbf8\ud130\ub97c \uc870\uc815\ud558\uace0, \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc744 \ud1b5\ud574 \ucd5c\uc801\uc758 \uc131\ub2a5\uc744 \ucc3e\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4.\n <\/p>\n5.5 \uc804\ub7b5 \uc2e4\ud589<\/h3>\n\n \ucd5c\uc801\uc758 \ubaa8\ub378\uc744 \ubc14\ud0d5\uc73c\ub85c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc744 \uc2e4\ud589\ud569\ub2c8\ub2e4.\n \uc774 \ub2e8\uacc4\uc5d0\uc11c\ub294 \uc2e4\uc2dc\uac04 \ub370\uc774\ud130 \uc2a4\ud2b8\ub9ac\ubc0d\uc744 \ubc1b\uc544 \ubaa8\ub378\uc744 \ud1b5\ud574 \ub9e4\uc218\/\ub9e4\ub3c4 \uc2e0\ud638\ub97c \ubc1c\uc0dd\uc2dc\ud0a4\uace0, \uc774\ub97c \uae30\ubc18\uc73c\ub85c \uc790\ub3d9\uc73c\ub85c \uac70\ub798\ub97c \uc218\ud589\ud558\ub294 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n6. \uacb0\ub860<\/h2>\n\n \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18 \uacb0\uc815 \uacfc\uc815\uc744 \ud1b5\ud574 \uc804\ud1b5\uc801\uc778 \ud2b8\ub808\uc774\ub529 \ubc29\uc2dd\uc5d0 \ube44\ud574\n \ub9ce\uc740 \uc774\uc810\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4. \ud2b9\ud788 LightGBM\uc740 \ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4\uc640 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc81c\uacf5\ud558\uc5ec,\n \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uac1c\ubc1c\uc5d0 \uc720\uc6a9\ud55c \ub3c4\uad6c\uc785\ub2c8\ub2e4. \uc9c0\uc18d\uc801\uc73c\ub85c \uc54c\uace0\ub9ac\uc998\uc744 \uac1c\uc120\ud558\uace0, \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\uc640 \uc804\ub7b5\uc744 \uc801\uc6a9\ud558\uc5ec\n \uc548\uc815\uc801\uc774\uace0 \uc218\uc775\uc131 \uc788\ub294 \ud2b8\ub808\uc774\ub529\uc744 \uc2e4\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n7. \ucc38\uace0 \uc790\ub8cc<\/h2>\n\nLightGBM Documentation<\/a><\/li>\nCoursera – Machine Learning by Andrew Ng<\/a><\/li>\nTowards Data Science<\/a> (\ub370\uc774\ud130 \uc0ac\uc774\uc5b8\uc2a4 \uad00\ub828 \ube14\ub85c\uadf8)<\/li>\n<\/ul>\n
\n ML4T\ub294 \uba38\uc2e0\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5 \uc218\ub9bd \ubc0f \ucd5c\uc801\ud654 \uc791\uc5c5\uc744 \uc758\ubbf8\ud569\ub2c8\ub2e4.\n \uc0ac\uc6a9\uc790\ub294 \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc95\uc744 \ud1b5\ud574 \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uace0, \uc774\ub97c \ud1b5\ud574 \ubcf4\ub2e4 \ud6a8\uacfc\uc801\uc778 \uc758\uc0ac\uacb0\uc815\uc744 \ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uae30 \uc704\ud574\uc11c\ub294 \uc6b0\uc120\uc801\uc73c\ub85c \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4.\n \uc8fc\uac00 \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud558\uae30 \uc704\ud574\uc11c\ub294 API\ub97c \uc0ac\uc6a9\ud558\uac70\ub098 \uae08\uc735 \ub370\uc774\ud130 \uc81c\uacf5 \uc5c5\uccb4\uc758 \ub3c4\uc6c0\uc744 \ubc1b\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n \uc218\uc9d1\ud55c \ub370\uc774\ud130\ub294 \ubaa8\ub378 \ud559\uc2b5\uc5d0 \uc801\ud569\ud55c \ud615\uc2dd\uc73c\ub85c \ubcc0\ud658\ub418\uc5b4\uc57c \ud569\ub2c8\ub2e4.\n \uc774 \uacfc\uc815\uc5d0\uc11c \uacb0\uce21\uce58\ub97c \ucc98\ub9ac\ud558\uace0, \ud2b9\uc131 \uc5d4\uc9c0\ub2c8\uc5b4\ub9c1\uc744 \ud1b5\ud574 \uae30\uc874 \ub370\uc774\ud130\ub85c\ubd80\ud130 \uc0c8\ub85c\uc6b4 \ud2b9\uc9d5\uc744 \uc0dd\uc131\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n \uc804\ucc98\ub9ac\ub41c \ub370\uc774\ud130\ub97c \uae30\ubc18\uc73c\ub85c LightGBM \ubaa8\ub378\uc744 \ud559\uc2b5\ud569\ub2c8\ub2e4.\n \uc544\ub798\ub294 Python\uc744 \uc0ac\uc6a9\ud55c LightGBM \ubaa8\ub378\uc758 \uae30\ubcf8\uc801\uc778 \ud559\uc2b5 \ucf54\ub4dc\uc785\ub2c8\ub2e4:\n <\/p>\n
\nimport lightgbm as lgb\nfrom sklearn.model_selection import train_test_split\n\n# \ub370\uc774\ud130\uc14b \ub85c\ub4dc\ndata = ... # \ub370\uc774\ud130 \ub85c\ub4dc \ucf54\ub4dc\nX = data.drop(columns='target')\ny = data['target']\n\n# \ud6c8\ub828 \ub370\uc774\ud130\uc640 \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub85c \ubd84\ud560\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\n# LightGBM \ub370\uc774\ud130\uc14b \uc0dd\uc131\ntrain_data = lgb.Dataset(X_train, label=y_train)\ntest_data = lgb.Dataset(X_test, label=y_test)\n\n# \ubaa8\ub378 \ud30c\ub77c\ubbf8\ud130 \uc124\uc815\nparams = {\n 'objective': 'binary',\n 'metric': 'auc',\n 'learning_rate': 0.05,\n 'num_leaves': 31,\n 'verbose': -1\n}\n\n# \ubaa8\ub378 \ud559\uc2b5\nmodel = lgb.train(params, train_data, num_boost_round=1000, valid_sets=[test_data], early_stopping_rounds=100)\n <\/code><\/pre>\n5.4 \ubaa8\ub378 \ud3c9\uac00<\/h3>\n\n \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uae30 \uc704\ud574 AUC(Area Under the Curve) \uc9c0\ud45c\ub97c \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n \ud3c9\uac00\ub41c \uc131\ub2a5\uc744 \uae30\ubc18\uc73c\ub85c \ubaa8\ub378\uc758 \ud30c\ub77c\ubbf8\ud130\ub97c \uc870\uc815\ud558\uace0, \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc744 \ud1b5\ud574 \ucd5c\uc801\uc758 \uc131\ub2a5\uc744 \ucc3e\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4.\n <\/p>\n5.5 \uc804\ub7b5 \uc2e4\ud589<\/h3>\n\n \ucd5c\uc801\uc758 \ubaa8\ub378\uc744 \ubc14\ud0d5\uc73c\ub85c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc744 \uc2e4\ud589\ud569\ub2c8\ub2e4.\n \uc774 \ub2e8\uacc4\uc5d0\uc11c\ub294 \uc2e4\uc2dc\uac04 \ub370\uc774\ud130 \uc2a4\ud2b8\ub9ac\ubc0d\uc744 \ubc1b\uc544 \ubaa8\ub378\uc744 \ud1b5\ud574 \ub9e4\uc218\/\ub9e4\ub3c4 \uc2e0\ud638\ub97c \ubc1c\uc0dd\uc2dc\ud0a4\uace0, \uc774\ub97c \uae30\ubc18\uc73c\ub85c \uc790\ub3d9\uc73c\ub85c \uac70\ub798\ub97c \uc218\ud589\ud558\ub294 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n6. \uacb0\ub860<\/h2>\n\n \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18 \uacb0\uc815 \uacfc\uc815\uc744 \ud1b5\ud574 \uc804\ud1b5\uc801\uc778 \ud2b8\ub808\uc774\ub529 \ubc29\uc2dd\uc5d0 \ube44\ud574\n \ub9ce\uc740 \uc774\uc810\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4. \ud2b9\ud788 LightGBM\uc740 \ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4\uc640 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc81c\uacf5\ud558\uc5ec,\n \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uac1c\ubc1c\uc5d0 \uc720\uc6a9\ud55c \ub3c4\uad6c\uc785\ub2c8\ub2e4. \uc9c0\uc18d\uc801\uc73c\ub85c \uc54c\uace0\ub9ac\uc998\uc744 \uac1c\uc120\ud558\uace0, \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\uc640 \uc804\ub7b5\uc744 \uc801\uc6a9\ud558\uc5ec\n \uc548\uc815\uc801\uc774\uace0 \uc218\uc775\uc131 \uc788\ub294 \ud2b8\ub808\uc774\ub529\uc744 \uc2e4\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n7. \ucc38\uace0 \uc790\ub8cc<\/h2>\n\nLightGBM Documentation<\/a><\/li>\nCoursera – Machine Learning by Andrew Ng<\/a><\/li>\nTowards Data Science<\/a> (\ub370\uc774\ud130 \uc0ac\uc774\uc5b8\uc2a4 \uad00\ub828 \ube14\ub85c\uadf8)<\/li>\n<\/ul>\n
\n \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uae30 \uc704\ud574 AUC(Area Under the Curve) \uc9c0\ud45c\ub97c \uc0ac\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n \ud3c9\uac00\ub41c \uc131\ub2a5\uc744 \uae30\ubc18\uc73c\ub85c \ubaa8\ub378\uc758 \ud30c\ub77c\ubbf8\ud130\ub97c \uc870\uc815\ud558\uace0, \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130 \ud29c\ub2dd\uc744 \ud1b5\ud574 \ucd5c\uc801\uc758 \uc131\ub2a5\uc744 \ucc3e\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4.\n <\/p>\n
\n \ucd5c\uc801\uc758 \ubaa8\ub378\uc744 \ubc14\ud0d5\uc73c\ub85c \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc744 \uc2e4\ud589\ud569\ub2c8\ub2e4.\n \uc774 \ub2e8\uacc4\uc5d0\uc11c\ub294 \uc2e4\uc2dc\uac04 \ub370\uc774\ud130 \uc2a4\ud2b8\ub9ac\ubc0d\uc744 \ubc1b\uc544 \ubaa8\ub378\uc744 \ud1b5\ud574 \ub9e4\uc218\/\ub9e4\ub3c4 \uc2e0\ud638\ub97c \ubc1c\uc0dd\uc2dc\ud0a4\uace0, \uc774\ub97c \uae30\ubc18\uc73c\ub85c \uc790\ub3d9\uc73c\ub85c \uac70\ub798\ub97c \uc218\ud589\ud558\ub294 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n
\n \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ub370\uc774\ud130 \uae30\ubc18 \uacb0\uc815 \uacfc\uc815\uc744 \ud1b5\ud574 \uc804\ud1b5\uc801\uc778 \ud2b8\ub808\uc774\ub529 \ubc29\uc2dd\uc5d0 \ube44\ud574\n \ub9ce\uc740 \uc774\uc810\uc744 \uc81c\uacf5\ud569\ub2c8\ub2e4. \ud2b9\ud788 LightGBM\uc740 \ube60\ub978 \ud559\uc2b5 \uc18d\ub3c4\uc640 \ub192\uc740 \uc815\ud655\ub3c4\ub97c \uc81c\uacf5\ud558\uc5ec,\n \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c \uac1c\ubc1c\uc5d0 \uc720\uc6a9\ud55c \ub3c4\uad6c\uc785\ub2c8\ub2e4. \uc9c0\uc18d\uc801\uc73c\ub85c \uc54c\uace0\ub9ac\uc998\uc744 \uac1c\uc120\ud558\uace0, \uc0c8\ub85c\uc6b4 \ub370\uc774\ud130\uc640 \uc804\ub7b5\uc744 \uc801\uc6a9\ud558\uc5ec\n \uc548\uc815\uc801\uc774\uace0 \uc218\uc775\uc131 \uc788\ub294 \ud2b8\ub808\uc774\ub529\uc744 \uc2e4\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n