\n \uc790\uc5f0\uc5b4 \ucc98\ub9ac(Natural Language Processing, NLP)\ub294 \ub525 \ub7ec\ub2dd\uc758 \ubc1c\uc804\uacfc \ud568\uaed8 \uae09\uc18d\ud788 \ubc1c\uc804\ud558\uace0 \uc788\ub294 \ubd84\uc57c\uc785\ub2c8\ub2e4. \uadf8 \uc911 \ud558\ub098\uc758 \uad6c\uccb4\uc801\uc778 \uc801\uc6a9 \uc0ac\ub840\ub85c \ub85c\uc774\ud130 \ub274\uc2a4 \ubd84\ub958\uac00 \uc788\uc2b5\ub2c8\ub2e4. \ubcf8 \uae00\uc5d0\uc11c\ub294 \ub85c\uc774\ud130 \ub274\uc2a4 \ub370\uc774\ud130\ub97c \uc774\uc6a9\ud558\uc5ec \ub274\uc2a4 \uae30\uc0ac\ub97c \ubd84\ub958\ud558\ub294 \ubc29\ubc95\uc744 \uc18c\uac1c\ud558\uace0, \ub525 \ub7ec\ub2dd \ubaa8\ub378\uc744 \ud65c\uc6a9\ud55c \uc790\uc5f0\uc5b4 \ucc98\ub9ac\uc758 \uae30\ucd08\ubd80\ud130 \uc2e4\uc2b5 \uc608\uc81c\uae4c\uc9c0 \uc0c1\uc138\ud788 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4.\n<\/p>\n
\n \uc790\uc5f0\uc5b4 \ucc98\ub9ac\ub294 \uc778\uac04\uc758 \uc5b8\uc5b4\ub97c \ucef4\ud4e8\ud130\uac00 \uc774\ud574\ud558\uace0 \ucc98\ub9ac\ud560 \uc218 \uc788\ub3c4\ub85d \ud558\ub294 \uae30\uc220\uc785\ub2c8\ub2e4. \uc790\uc5f0\uc5b4 \ucc98\ub9ac\ub294 \ud14d\uc2a4\ud2b8 \ubd84\uc11d, \uae30\uacc4 \ubc88\uc5ed, \uac10\uc815 \ubd84\uc11d \ub4f1 \ub2e4\uc591\ud55c \ubd84\uc57c\uc5d0 \ud65c\uc6a9\ub418\uba70, \ucd5c\uadfc\uc5d0\ub294 \ub525 \ub7ec\ub2dd \uae30\uc220\uc758 \ubc1c\uc804\uc73c\ub85c \ub354\uc6b1 \uc815\ud655\ud558\uace0 \ud6a8\uc728\uc801\uc778 \uacb0\uacfc\ub97c \uc5bb\uace0 \uc788\uc2b5\ub2c8\ub2e4.\n<\/p>\n
\n \ub85c\uc774\ud130 \ub274\uc2a4 \ub370\uc774\ud130\uc14b\uc740 1986\ub144 \ub85c\uc774\ud130\uc5d0\uc11c \uc218\uc9d1\ud55c \ub274\uc2a4 \uae30\uc0ac \ubaa8\uc74c\uc73c\ub85c, \ub274\uc2a4 \uae30\uc0ac\ub97c \uc5ec\ub7ec \uce74\ud14c\uace0\ub9ac\ub85c \ubd84\ub958\ud558\ub294 \uc791\uc5c5\uc5d0 \uc720\uc6a9\ud569\ub2c8\ub2e4. \uc774 \ub370\uc774\ud130\uc14b\uc740 \uce74\ud14c\uace0\ub9ac\uac00 90\uac1c\ub85c \ub098\ub204\uc5b4\uc838 \uc788\uc73c\uba70, \uac01 \uce74\ud14c\uace0\ub9ac\uc5d0\ub294 \ub2e4\uc218\uc758 \ub274\uc2a4 \uae30\uc0ac\uac00 \ud3ec\ud568\ub418\uc5b4 \uc788\uc2b5\ub2c8\ub2e4. \ub85c\uc774\ud130 \ub370\uc774\ud130\uc14b\uc740 \uc624\ub298\ub0a0 \ub2e4\uc591\ud55c \ud14d\uc2a4\ud2b8 \ubd84\ub958 \uc5f0\uad6c\uc5d0\uc11c \ub110\ub9ac \uc0ac\uc6a9\ub418\uace0 \uc788\uc2b5\ub2c8\ub2e4.\n<\/p>\n
\n \ub85c\uc774\ud130 \ub274\uc2a4 \ub370\uc774\ud130\uc14b\uc740 \ubcf4\ud1b5 \ud6c8\ub828 \ub370\uc774\ud130\uc640 \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub85c \ub098\ub204\uc5b4\uc9d1\ub2c8\ub2e4. \uac01 \ub274\uc2a4 \uae30\uc0ac\ub294 \ub2e4\uc74c\uacfc \uac19\uc740 \uc815\ubcf4\ub85c \uad6c\uc131\ub429\ub2c8\ub2e4:\n<\/p>\n
\n \ubaa8\ub378 \ud6c8\ub828\uc744 \uc704\ud574\uc11c\ub294 \ub370\uc774\ud130 \uc804\ucc98\ub9ac\uac00 \ud544\uc218\uc801\uc785\ub2c8\ub2e4. \uc5ec\uae30\uc11c\ub294 \ud30c\uc774\uc36c\uc744 \uc0ac\uc6a9\ud558\uc5ec \ub370\uc774\ud130\ub97c \ub85c\ub4dc\ud558\uace0 \uc804\ucc98\ub9ac\ud558\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud558\uaca0\uc2b5\ub2c8\ub2e4. \ub370\uc774\ud130 \uc804\ucc98\ub9ac \uacfc\uc815\uc740 \uc8fc\ub85c \ub2e4\uc74c\uacfc \uac19\uc740 \ub2e8\uacc4\ub85c \uc774\ub8e8\uc5b4\uc9d1\ub2c8\ub2e4:\n<\/p>\n
\nimport pandas as pd\n\n# \ub85c\uc774\ud130 \ub274\uc2a4 \ub370\uc774\ud130\uc14b \ub85c\ub4dc\ndataframe = pd.read_csv('reuters.csv') # \ub370\uc774\ud130\uc14b \uacbd\ub85c\nprint(dataframe.head())\n<\/code><\/pre>\n3.2 \ud14d\uc2a4\ud2b8 \uc815\uc81c \ubc0f \uc804\ucc98\ub9ac<\/h3>\n
\n \ub274\uc2a4 \uae30\uc0ac\ub294 \uc885\uc885 \ubd88\ud544\uc694\ud55c \ubb38\uc790\ub098 \uae30\ud638\ub97c \ud3ec\ud568\ud560 \uc218 \uc788\uc73c\ubbc0\ub85c, \uc815\uc81c \uacfc\uc815\uc774 \ud544\uc694\ud569\ub2c8\ub2e4. \uc77c\ubc18\uc801\uc73c\ub85c \uc0ac\uc6a9\ub418\ub294 \uc815\uc81c \uc791\uc5c5\uc740 \ub2e4\uc74c\uacfc \uac19\uc2b5\ub2c8\ub2e4:\n<\/p>\n
\n- \ud2b9\uc218 \ubb38\uc790 \uc81c\uac70<\/li>\n
- \uc18c\ubb38\uc790 \ubcc0\ud658<\/li>\n
- \ubd88\uc6a9\uc5b4(stop words) \uc81c\uac70<\/li>\n
- \uc5b4\uac04 \ucd94\ucd9c(stemming) \ub610\ub294 \ud45c\uc81c\uc5b4 \ucd94\ucd9c(lemmatization)<\/li>\n<\/ul>\n
\nimport re\nfrom nltk.corpus import stopwords\nfrom nltk.stem import PorterStemmer\n\n# \uc815\uc81c \ud568\uc218 \uc815\uc758\ndef clean_text(text):\n text = re.sub(r'\\W', ' ', text) # \ud2b9\uc218 \ubb38\uc790 \uc81c\uac70\n text = text.lower() # \uc18c\ubb38\uc790 \ubcc0\ud658\n text = ' '.join([word for word in text.split() if word not in stopwords.words('english')])\n return text\n\n# \ub370\uc774\ud130\ud504\ub808\uc784\uc5d0\uc11c \ud14d\uc2a4\ud2b8 \uc815\uc81c\ndataframe['cleaned_text'] = dataframe['text'].apply(clean_text)\n<\/code><\/pre>\n4. \ub525 \ub7ec\ub2dd \ubaa8\ub378 \uad6c\ucd95<\/h2>\n
\n \uc804\ucc98\ub9ac\ub41c \ub370\uc774\ud130\ub97c \uc0ac\uc6a9\ud558\uc5ec \ub525 \ub7ec\ub2dd \ubaa8\ub378\uc744 \uad6c\ucd95\ud558\uaca0\uc2b5\ub2c8\ub2e4. \uc77c\ubc18\uc801\uc73c\ub85c \ud14d\uc2a4\ud2b8 \ubd84\ub958\ub97c \uc704\ud574 \uc21c\ud658 \uc2e0\uacbd\ub9dd(RNN) \ub610\ub294 \uadf8 \ubcc0\ud615\uc778 LSTM(Long Short-Term Memory) \ubaa8\ub378\uc744 \uc0ac\uc6a9\ud569\ub2c8\ub2e4. \uc5ec\uae30\uc11c\ub294 Keras\ub97c \uc774\uc6a9\ud558\uc5ec LSTM \ubaa8\ub378\uc744 \uad6c\ud604\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.\n<\/p>\n
4.1 \ubaa8\ub378 \uc124\uacc4<\/h3>\n\nfrom keras.models import Sequential\nfrom keras.layers import Dense, Embedding, LSTM, SpatialDropout1D\n\n# \ud30c\ub77c\ubbf8\ud130 \uc124\uc815\nembedding_dim = 100\nmax_length = 200\nmodel = Sequential()\nmodel.add(Embedding(input_dim=vocab_size, output_dim=embedding_dim, input_length=max_length))\nmodel.add(SpatialDropout1D(0.2))\nmodel.add(LSTM(100, dropout=0.2, recurrent_dropout=0.2))\nmodel.add(Dense(num_classes, activation='softmax'))\n\nmodel.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n<\/code><\/pre>\n4.2 \ubaa8\ub378 \ud6c8\ub828<\/h3>\n\nhistory = model.fit(X_train, y_train, epochs=5, batch_size=64, validation_data=(X_test, y_test), verbose=1)\n<\/code><\/pre>\n5. \ubaa8\ub378 \ud3c9\uac00 \ubc0f \uacb0\uacfc \ubd84\uc11d<\/h2>\n
\n \ud6c8\ub828\uc774 \uc644\ub8cc\ub41c \ud6c4, \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud569\ub2c8\ub2e4. \uc77c\ubc18\uc801\uc73c\ub85c \uc815\ubc00\ub3c4, \uc7ac\ud604\uc728, F1 \uc810\uc218\ub97c \uae30\uc900\uc73c\ub85c \ubaa8\ub378 \uc131\ub2a5\uc744 \uce21\uc815\ud569\ub2c8\ub2e4.\n<\/p>\n
\nfrom sklearn.metrics import classification_report\n\ny_pred = model.predict(X_test)\nprint(classification_report(y_test, y_pred))\n<\/code><\/pre>\n6. \uacb0\ub860<\/h2>\n
\n \ubcf8 \ud3ec\uc2a4\ud305\uc5d0\uc11c\ub294 \ub525 \ub7ec\ub2dd\uc744 \uc774\uc6a9\ud55c \uc790\uc5f0\uc5b4 \ucc98\ub9ac\uc758 \uae30\ubcf8 \uac1c\ub150\uacfc \ub85c\uc774\ud130 \ub274\uc2a4 \ubd84\ub958\ub97c \uc704\ud55c \uc804\ucc98\ub9ac, \ubaa8\ub378 \uad6c\ucd95 \ubc0f \ud3c9\uac00 \ubc29\ubc95\uc5d0 \ub300\ud574 \uc0b4\ud3b4\ubcf4\uc558\uc2b5\ub2c8\ub2e4. \uc774\ub7ec\ud55c \uacfc\uc815\ub4e4\uc744 \ud1b5\ud574 \ub525 \ub7ec\ub2dd \uae30\ubc18\uc758 \uc790\uc5f0\uc5b4 \ucc98\ub9ac \ubaa8\ub378\uc744 \uad6c\ucd95\ud558\uace0, \uc2e4\uc81c\uc801\uc778 \ub370\uc774\ud130 \ubd84\uc11d \ubc0f \ubd84\ub958 \uc791\uc5c5\uc744 \uc218\ud589\ud560 \uc218 \uc788\ub294 \uae30\ubc18\uc744 \ub2e4\uc84c\uc2b5\ub2c8\ub2e4.\n<\/p>\n
7. \ucc38\uace0 \ubb38\ud5cc<\/h2>\n