# Update 2021.04.26 ## 作法 CensorWords_Ground Truth_precise_210419_1316.csv 拿去斷詞 predict ## 結果 **CensorWords_Ground Truth_precise_210419_1316.csv** ``` 斷詞前 Predict Accuracy : 0.829118460120815 Model correct 0 1 2 3 4 5 predict 0 12877 692 34 1337 432 85 1 8 1923 2 166 14 0 2 2 2 108 12 55 0 3 1 31 0 446 5 0 4 6 5 12 123 746 2 5 1 1 2 304 4 96 ``` ``` CensorWords_Ground Truth: 19534 0 12895 1 2654 3 2388 4 1256 5 183 2 158 Label Accuracy : 0.8237432169550527 數美 correct 0 1 2 3 4 5 label 0 12894 1116 59 1125 801 179 1 1 1536 1 96 0 0 2 0 0 91 0 27 0 3 0 2 0 1146 6 0 4 0 0 7 20 421 1 5 0 0 0 1 1 3 Predict Accuracy : 0.8374116924337053 Model correct 0 1 2 3 4 5 predict 0 12753 911 65 987 450 111 1 92 1703 2 130 14 1 2 3 1 79 0 41 0 3 29 35 1 1020 14 0 4 14 3 9 29 734 2 5 4 1 2 222 3 69 ``` https://docs.google.com/spreadsheets/d/1E_jU-_UOfK1xrCzh0k7mrVtD28-bhn544JED-BsMdWU/edit?usp=sharing --- # Ground truth ## Train: new word dataset_add label_normal name_default ## Test: Ground truth ## Model text_cnn_best_99.47767857142857_LR0.001_BATCH100_EPOCH100 ## result(有斷詞) ``` CensorWords_Ground Truth_precise_210425_1810.csv Predict 0 1 2 3 4 5 All correct 0 12877 8 2 1 6 1 12895 1 692 1923 2 31 5 1 2654 2 34 2 108 0 12 2 158 3 1337 166 12 446 123 304 2388 4 432 14 55 5 746 4 1256 5 85 0 0 0 2 96 183 All 15457 2113 179 483 894 408 19534 Accuracy: 0.829118460120815 Error amount: 3338 / 19534 Type 0 Accuracy: 12877 / 12895 = 0.998604 Type 1 Accuracy: 1923 / 2654 = 0.724567 Type 2 Accuracy: 108 / 158 = 0.683544 Type 3 Accuracy: 446 / 2388 = 0.186767 Type 4 Accuracy: 746 / 1256 = 0.593949 Type 5 Accuracy: 96 / 183 = 0.524590 ``` ## result (沒斷詞) ``` predict correct number: 15190 accuracy: 0.7781762295081968 predict 0 1 2 3 4 5 correct 0 12500 274 12 7 92 9 1 869 1726 3 21 28 0 2 23 4 110 1 19 1 3 1573 279 13 180 269 68 4 505 47 55 8 640 1 5 136 6 1 1 5 34 ``` --- # ULR detect ### 作法 ![](https://i.imgur.com/UmbbfxH.png) https://www.notion.so/0ea0785783454d2fb05b7afc2f0d7401?v=4ac201766ff1426eb514b67750643558&p=b6ffd73c1c1a4404bfe56b539033c99a ## 結果 ``` Predict 0 4 5 All label 5 56 2 5728 5786 All 56 2 5728 5786 Accuracy: 0.9899758036640166 Error amount: 58 / 5786 Type 0 Accuracy: 56 / 5786 = 0.009679 ``` ``` /content/drive/Shared drives/DataScience/1091_Spam_Detect/nickname_predict_label_result/CensorWords_Ground Truth_precise_210427_1723.csv Predict 0 1 2 3 4 5 All label 0 13368 1171 43 635 744 213 16174 1 263 1298 0 69 2 2 1634 2 18 0 67 0 32 1 118 3 201 63 0 761 25 104 1154 4 127 10 13 6 292 1 449 5 3 0 0 0 2 0 5 All 13980 2542 123 1471 1097 321 19534 斷詞 Accuracy: 0.8081294153783147 Error amount: 3748 / 19534 Type 0 Accuracy: 13368 / 16174 = 0.826512 Type 1 Accuracy: 1298 / 1634 = 0.794370 Type 2 Accuracy: 67 / 118 = 0.567797 Type 3 Accuracy: 761 / 1154 = 0.659445 Type 4 Accuracy: 292 / 449 = 0.650334 Type 5 Accuracy: 0 / 5 = 0.000000 ```