Page MenuHomePhabricator

Test draftquality sentiment feature on Editquality
Closed, ResolvedPublic

Description

To generate a tuning report of editquality on enwiki using the sentiment( https://github.com/wiki-ai/draftquality/pull/9 ) feature and test its viability.

Event Timeline

Enwiki damaging gives slight rise in accuracy:

Model tuning report

  • Revscoring version: 1.3.17
  • Features: editquality.feature_lists.enwiki.damaging
  • Date: 2017-07-20T02:42:12.884370
  • Observations: 19460
  • Labels: [false, true]
  • Scoring: roc_auc
  • Folds: 5

Top scoring configurations

modelmean(scores)std(scores)params
:------------------------------------------:--------------::-----------------------------------------------------------------------
GradientBoostingClassifier0.9220.007n_estimators=700, max_depth=5, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.9210.007n_estimators=500, max_depth=5, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.9210.009n_estimators=700, max_depth=7, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.920.008n_estimators=500, max_depth=7, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.9190.005n_estimators=100, max_depth=3, learning_rate=0.1, max_features="log2"
GradientBoostingClassifier0.9190.009n_estimators=300, max_depth=3, learning_rate=0.1, max_features="log2"
GradientBoostingClassifier0.9190.006n_estimators=700, max_depth=3, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.9180.008n_estimators=300, max_depth=7, learning_rate=0.01, max_features="log2"
GradientBoostingClassifier0.9180.007n_estimators=700, max_depth=1, learning_rate=0.1, max_features="log2"
GradientBoostingClassifier0.9180.006n_estimators=300, max_depth=5, learning_rate=0.01, max_features="log2"

Models

GradientBoostingClassifier

mean(scores)std(scores)params
---------------:--------------::-----------------------------------------------------------------------
0.9220.007n_estimators=700, max_depth=5, learning_rate=0.01, max_features="log2"
0.9210.007n_estimators=500, max_depth=5, learning_rate=0.01, max_features="log2"
0.9210.009n_estimators=700, max_depth=7, learning_rate=0.01, max_features="log2"
0.920.008n_estimators=500, max_depth=7, learning_rate=0.01, max_features="log2"
0.9190.005n_estimators=100, max_depth=3, learning_rate=0.1, max_features="log2"
0.9190.009n_estimators=300, max_depth=3, learning_rate=0.1, max_features="log2"
0.9190.006n_estimators=700, max_depth=3, learning_rate=0.01, max_features="log2"
0.9180.008n_estimators=300, max_depth=7, learning_rate=0.01, max_features="log2"
0.9180.007n_estimators=700, max_depth=1, learning_rate=0.1, max_features="log2"
0.9180.006n_estimators=300, max_depth=5, learning_rate=0.01, max_features="log2"
0.9180.008n_estimators=100, max_depth=5, learning_rate=0.1, max_features="log2"
0.9170.007n_estimators=500, max_depth=1, learning_rate=0.1, max_features="log2"
0.9170.006n_estimators=300, max_depth=1, learning_rate=0.5, max_features="log2"
0.9170.012n_estimators=500, max_depth=3, learning_rate=0.1, max_features="log2"
0.9160.006n_estimators=500, max_depth=3, learning_rate=0.01, max_features="log2"
0.9160.009n_estimators=500, max_depth=1, learning_rate=0.5, max_features="log2"
0.9150.006n_estimators=300, max_depth=1, learning_rate=0.1, max_features="log2"
0.9140.008n_estimators=100, max_depth=1, learning_rate=0.5, max_features="log2"
0.9140.01n_estimators=700, max_depth=3, learning_rate=0.1, max_features="log2"
0.9140.006n_estimators=100, max_depth=7, learning_rate=0.01, max_features="log2"
0.9140.009n_estimators=700, max_depth=1, learning_rate=0.5, max_features="log2"
0.9130.006n_estimators=300, max_depth=3, learning_rate=0.01, max_features="log2"
0.9120.012n_estimators=300, max_depth=5, learning_rate=0.1, max_features="log2"
0.9120.006n_estimators=100, max_depth=5, learning_rate=0.01, max_features="log2"
0.9120.008n_estimators=100, max_depth=7, learning_rate=0.1, max_features="log2"
0.9090.012n_estimators=700, max_depth=5, learning_rate=0.1, max_features="log2"
0.9080.011n_estimators=500, max_depth=5, learning_rate=0.1, max_features="log2"
0.9070.013n_estimators=300, max_depth=7, learning_rate=0.1, max_features="log2"
0.9070.011n_estimators=700, max_depth=7, learning_rate=0.1, max_features="log2"
0.9050.011n_estimators=500, max_depth=7, learning_rate=0.1, max_features="log2"
0.9030.008n_estimators=100, max_depth=3, learning_rate=0.01, max_features="log2"
0.9030.007n_estimators=100, max_depth=1, learning_rate=0.1, max_features="log2"
0.9020.007n_estimators=700, max_depth=1, learning_rate=0.01, max_features="log2"
0.8970.021n_estimators=100, max_depth=1, learning_rate=1, max_features="log2"
0.8950.007n_estimators=500, max_depth=1, learning_rate=0.01, max_features="log2"
0.8920.013n_estimators=100, max_depth=3, learning_rate=0.5, max_features="log2"
0.8920.013n_estimators=300, max_depth=3, learning_rate=0.5, max_features="log2"
0.8920.016n_estimators=500, max_depth=1, learning_rate=1, max_features="log2"
0.8860.011n_estimators=300, max_depth=1, learning_rate=0.01, max_features="log2"
0.8850.03n_estimators=300, max_depth=1, learning_rate=1, max_features="log2"
0.8830.017n_estimators=700, max_depth=5, learning_rate=0.5, max_features="log2"
0.8820.016n_estimators=500, max_depth=3, learning_rate=0.5, max_features="log2"
0.8810.019n_estimators=700, max_depth=1, learning_rate=1, max_features="log2"
0.8760.026n_estimators=700, max_depth=3, learning_rate=0.5, max_features="log2"
0.8750.018n_estimators=500, max_depth=5, learning_rate=0.5, max_features="log2"
0.8710.014n_estimators=100, max_depth=1, learning_rate=0.01, max_features="log2"
0.8670.021n_estimators=500, max_depth=7, learning_rate=0.5, max_features="log2"
0.8620.021n_estimators=100, max_depth=5, learning_rate=0.5, max_features="log2"
0.8560.046n_estimators=300, max_depth=5, learning_rate=0.5, max_features="log2"
0.8470.06n_estimators=100, max_depth=7, learning_rate=0.5, max_features="log2"
0.8370.05n_estimators=100, max_depth=3, learning_rate=1, max_features="log2"
0.8360.092n_estimators=300, max_depth=7, learning_rate=0.5, max_features="log2"
0.8350.072n_estimators=700, max_depth=7, learning_rate=0.5, max_features="log2"
0.8060.016n_estimators=100, max_depth=5, learning_rate=1, max_features="log2"
0.7790.055n_estimators=300, max_depth=3, learning_rate=1, max_features="log2"
0.7760.06n_estimators=500, max_depth=3, learning_rate=1, max_features="log2"
0.7130.066n_estimators=100, max_depth=7, learning_rate=1, max_features="log2"
0.7110.101n_estimators=300, max_depth=5, learning_rate=1, max_features="log2"
0.6960.154n_estimators=700, max_depth=3, learning_rate=1, max_features="log2"
0.6820.073n_estimators=300, max_depth=7, learning_rate=1, max_features="log2"
0.6450.118n_estimators=700, max_depth=5, learning_rate=1, max_features="log2"
0.580.152n_estimators=500, max_depth=5, learning_rate=1, max_features="log2"
0.5590.182n_estimators=700, max_depth=7, learning_rate=1, max_features="log2"
0.4670.123n_estimators=500, max_depth=7, learning_rate=1, max_features="log2"

GaussianNB

mean(scores)std(scores)params
---------------:--------------::---------
0.7770.03

LogisticRegression

mean(scores)std(scores)params
---------------:--------------::--------------------
0.9030.004penalty="l1", C=10
0.9030.005penalty="l1", C=1
0.8960.008penalty="l1", C=0.1
0.8460.007penalty="l2", C=1
0.8450.011penalty="l2", C=0.1
0.8420.01penalty="l2", C=10

BernoulliNB

mean(scores)std(scores)params
---------------:--------------::---------
0.8280.012

RandomForestClassifier

mean(scores)std(scores)params
---------------:--------------::--------------------------------------------------------------------------------
0.9170.006criterion="entropy", n_estimators=640, min_samples_leaf=13, max_features="log2"
0.9170.005criterion="entropy", n_estimators=160, min_samples_leaf=7, max_features="log2"
0.9160.006criterion="entropy", n_estimators=640, min_samples_leaf=7, max_features="log2"
0.9160.005criterion="entropy", n_estimators=320, min_samples_leaf=13, max_features="log2"
0.9160.006criterion="entropy", n_estimators=640, min_samples_leaf=5, max_features="log2"
0.9150.007criterion="entropy", n_estimators=320, min_samples_leaf=5, max_features="log2"
0.9150.007criterion="entropy", n_estimators=160, min_samples_leaf=13, max_features="log2"
0.9150.006criterion="entropy", n_estimators=320, min_samples_leaf=7, max_features="log2"
0.9140.007criterion="entropy", n_estimators=640, min_samples_leaf=3, max_features="log2"
0.9140.006criterion="entropy", n_estimators=320, min_samples_leaf=3, max_features="log2"
0.9140.006criterion="entropy", n_estimators=80, min_samples_leaf=13, max_features="log2"
0.9130.008criterion="entropy", n_estimators=160, min_samples_leaf=5, max_features="log2"
0.9120.006criterion="gini", n_estimators=160, min_samples_leaf=13, max_features="log2"
0.9120.006criterion="gini", n_estimators=640, min_samples_leaf=13, max_features="log2"
0.9120.004criterion="entropy", n_estimators=640, min_samples_leaf=1, max_features="log2"
0.9120.006criterion="gini", n_estimators=640, min_samples_leaf=7, max_features="log2"
0.9120.006criterion="gini", n_estimators=320, min_samples_leaf=13, max_features="log2"
0.9110.006criterion="entropy", n_estimators=80, min_samples_leaf=7, max_features="log2"
0.9110.006criterion="gini", n_estimators=640, min_samples_leaf=3, max_features="log2"
0.9110.006criterion="gini", n_estimators=640, min_samples_leaf=5, max_features="log2"
0.9110.005criterion="gini", n_estimators=320, min_samples_leaf=7, max_features="log2"
0.9110.004criterion="gini", n_estimators=320, min_samples_leaf=5, max_features="log2"
0.910.008criterion="entropy", n_estimators=160, min_samples_leaf=3, max_features="log2"
0.910.007criterion="entropy", n_estimators=40, min_samples_leaf=13, max_features="log2"
0.910.006criterion="gini", n_estimators=160, min_samples_leaf=7, max_features="log2"
0.9090.004criterion="entropy", n_estimators=80, min_samples_leaf=5, max_features="log2"
0.9080.005criterion="gini", n_estimators=320, min_samples_leaf=3, max_features="log2"
0.9080.007criterion="gini", n_estimators=40, min_samples_leaf=13, max_features="log2"
0.9080.004criterion="entropy", n_estimators=320, min_samples_leaf=1, max_features="log2"
0.9080.006criterion="gini", n_estimators=80, min_samples_leaf=13, max_features="log2"
0.9080.007criterion="gini", n_estimators=160, min_samples_leaf=5, max_features="log2"
0.9070.005criterion="gini", n_estimators=80, min_samples_leaf=5, max_features="log2"
0.9070.008criterion="gini", n_estimators=80, min_samples_leaf=7, max_features="log2"
0.9070.006criterion="entropy", n_estimators=40, min_samples_leaf=7, max_features="log2"
0.9060.006criterion="entropy", n_estimators=40, min_samples_leaf=5, max_features="log2"
0.9050.006criterion="gini", n_estimators=160, min_samples_leaf=3, max_features="log2"
0.9040.008criterion="gini", n_estimators=640, min_samples_leaf=1, max_features="log2"
0.9040.005criterion="gini", n_estimators=320, min_samples_leaf=1, max_features="log2"
0.9040.006criterion="entropy", n_estimators=160, min_samples_leaf=1, max_features="log2"
0.9030.008criterion="gini", n_estimators=40, min_samples_leaf=7, max_features="log2"
0.9030.005criterion="gini", n_estimators=40, min_samples_leaf=5, max_features="log2"
0.9020.008criterion="entropy", n_estimators=80, min_samples_leaf=3, max_features="log2"
0.9010.003criterion="entropy", n_estimators=40, min_samples_leaf=3, max_features="log2"
0.9010.009criterion="entropy", n_estimators=20, min_samples_leaf=13, max_features="log2"
0.9010.004criterion="gini", n_estimators=160, min_samples_leaf=1, max_features="log2"
0.90.006criterion="gini", n_estimators=80, min_samples_leaf=3, max_features="log2"
0.8990.006criterion="gini", n_estimators=40, min_samples_leaf=3, max_features="log2"
0.8990.009criterion="gini", n_estimators=20, min_samples_leaf=7, max_features="log2"
0.8970.009criterion="entropy", n_estimators=20, min_samples_leaf=7, max_features="log2"
0.8970.007criterion="entropy", n_estimators=80, min_samples_leaf=1, max_features="log2"
0.8960.009criterion="gini", n_estimators=20, min_samples_leaf=13, max_features="log2"
0.8950.008criterion="entropy", n_estimators=20, min_samples_leaf=5, max_features="log2"
0.8940.005criterion="gini", n_estimators=20, min_samples_leaf=5, max_features="log2"
0.890.009criterion="entropy", n_estimators=10, min_samples_leaf=13, max_features="log2"
0.890.01criterion="gini", n_estimators=10, min_samples_leaf=13, max_features="log2"
0.8880.005criterion="gini", n_estimators=80, min_samples_leaf=1, max_features="log2"
0.8830.009criterion="entropy", n_estimators=20, min_samples_leaf=3, max_features="log2"
0.8830.019criterion="gini", n_estimators=10, min_samples_leaf=7, max_features="log2"
0.8830.006criterion="gini", n_estimators=20, min_samples_leaf=3, max_features="log2"
0.880.005criterion="entropy", n_estimators=10, min_samples_leaf=7, max_features="log2"
0.8790.002criterion="gini", n_estimators=40, min_samples_leaf=1, max_features="log2"
0.8760.012criterion="entropy", n_estimators=10, min_samples_leaf=5, max_features="log2"
0.8740.015criterion="gini", n_estimators=10, min_samples_leaf=5, max_features="log2"
0.8730.015criterion="entropy", n_estimators=40, min_samples_leaf=1, max_features="log2"
0.8590.013criterion="entropy", n_estimators=10, min_samples_leaf=3, max_features="log2"
0.8520.021criterion="entropy", n_estimators=20, min_samples_leaf=1, max_features="log2"
0.8460.004criterion="gini", n_estimators=10, min_samples_leaf=3, max_features="log2"
0.8440.011criterion="gini", n_estimators=20, min_samples_leaf=1, max_features="log2"
0.8060.009criterion="entropy", n_estimators=10, min_samples_leaf=1, max_features="log2"
0.8020.012criterion="gini", n_estimators=10, min_samples_leaf=1, max_features="log2"

Goodfaith model shows a slight fall in accuracy:

Model tuning report

  • Revscoring version: 1.3.17
  • Features: editquality.feature_lists.enwiki.goodfaith
  • Date: 2017-07-20T15:34:07.565220
  • Observations: 19460
  • Labels: [false, true]
  • Scoring: roc_auc
  • Folds: 5

Top scoring configurations

modelmean(scores)std(scores)params
:------------------------------------------:--------------::-----------------------------------------------------------------------
GradientBoostingClassifier0.9070.012max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=700
GradientBoostingClassifier0.9060.014max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=500
GradientBoostingClassifier0.9050.013max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=500
GradientBoostingClassifier0.9050.011max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=700
GradientBoostingClassifier0.9040.015max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=700
GradientBoostingClassifier0.9040.014max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=300
GradientBoostingClassifier0.9020.012max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=300
GradientBoostingClassifier0.9020.011max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=700
GradientBoostingClassifier0.9020.011max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=500
GradientBoostingClassifier0.9020.011max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=500

Models

LogisticRegression

mean(scores)std(scores)params
---------------:--------------::--------------------
0.8870.011C=1, penalty="l1"
0.8860.012C=10, penalty="l1"
0.8810.014C=0.1, penalty="l1"
0.8360.017C=1, penalty="l2"
0.8290.008C=10, penalty="l2"
0.8280.017C=0.1, penalty="l2"

GradientBoostingClassifier

mean(scores)std(scores)params
---------------:--------------::-----------------------------------------------------------------------
0.9070.012max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=700
0.9060.014max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=500
0.9050.013max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=500
0.9050.011max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=700
0.9040.015max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=700
0.9040.014max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=300
0.9020.012max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=300
0.9020.011max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=700
0.9020.011max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=500
0.9020.011max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=500
0.9010.014max_features="log2", max_depth=3, learning_rate=0.1, n_estimators=100
0.90.011max_features="log2", max_depth=3, learning_rate=0.1, n_estimators=300
0.90.016max_features="log2", max_depth=5, learning_rate=0.1, n_estimators=100
0.90.01max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=300
0.8990.012max_features="log2", max_depth=1, learning_rate=0.5, n_estimators=500
0.8990.014max_features="log2", max_depth=7, learning_rate=0.01, n_estimators=100
0.8980.015max_features="log2", max_depth=5, learning_rate=0.1, n_estimators=300
0.8970.015max_features="log2", max_depth=3, learning_rate=0.1, n_estimators=500
0.8970.01max_features="log2", max_depth=1, learning_rate=0.5, n_estimators=700
0.8970.009max_features="log2", max_depth=1, learning_rate=0.5, n_estimators=300
0.8970.015max_features="log2", max_depth=7, learning_rate=0.1, n_estimators=100
0.8960.011max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=300
0.8950.014max_features="log2", max_depth=5, learning_rate=0.01, n_estimators=100
0.8940.017max_features="log2", max_depth=3, learning_rate=0.1, n_estimators=700
0.8940.014max_features="log2", max_depth=1, learning_rate=0.5, n_estimators=100
0.8910.02max_features="log2", max_depth=7, learning_rate=0.1, n_estimators=300
0.8910.02max_features="log2", max_depth=5, learning_rate=0.1, n_estimators=500
0.8890.01max_features="log2", max_depth=1, learning_rate=0.1, n_estimators=100
0.8890.018max_features="log2", max_depth=5, learning_rate=0.1, n_estimators=700
0.8870.021max_features="log2", max_depth=7, learning_rate=0.1, n_estimators=700
0.8870.011max_features="log2", max_depth=1, learning_rate=0.01, n_estimators=700
0.8860.015max_features="log2", max_depth=3, learning_rate=0.01, n_estimators=100
0.8850.022max_features="log2", max_depth=7, learning_rate=0.1, n_estimators=500
0.880.014max_features="log2", max_depth=1, learning_rate=0.01, n_estimators=500
0.8750.014max_features="log2", max_depth=3, learning_rate=0.5, n_estimators=100
0.8690.016max_features="log2", max_depth=1, learning_rate=0.01, n_estimators=300
0.8660.017max_features="log2", max_depth=3, learning_rate=0.5, n_estimators=700
0.8630.018max_features="log2", max_depth=3, learning_rate=0.5, n_estimators=300
0.8610.019max_features="log2", max_depth=3, learning_rate=0.5, n_estimators=500
0.8610.016max_features="log2", max_depth=1, learning_rate=0.01, n_estimators=100
0.860.015max_features="log2", max_depth=5, learning_rate=0.5, n_estimators=700
0.8560.024max_features="log2", max_depth=1, learning_rate=1, n_estimators=700
0.8490.025max_features="log2", max_depth=5, learning_rate=0.5, n_estimators=500
0.8470.02max_features="log2", max_depth=7, learning_rate=0.5, n_estimators=500
0.8470.021max_features="log2", max_depth=7, learning_rate=0.5, n_estimators=700
0.8440.027max_features="log2", max_depth=5, learning_rate=0.5, n_estimators=300
0.8410.044max_features="log2", max_depth=1, learning_rate=1, n_estimators=300
0.8360.05max_features="log2", max_depth=7, learning_rate=0.5, n_estimators=300
0.8360.027max_features="log2", max_depth=7, learning_rate=0.5, n_estimators=100
0.8350.023max_features="log2", max_depth=3, learning_rate=1, n_estimators=100
0.8320.016max_features="log2", max_depth=5, learning_rate=0.5, n_estimators=100
0.8290.046max_features="log2", max_depth=1, learning_rate=1, n_estimators=100
0.8230.048max_features="log2", max_depth=1, learning_rate=1, n_estimators=500
0.7610.048max_features="log2", max_depth=3, learning_rate=1, n_estimators=300
0.7520.073max_features="log2", max_depth=3, learning_rate=1, n_estimators=500
0.7450.029max_features="log2", max_depth=3, learning_rate=1, n_estimators=700
0.7370.05max_features="log2", max_depth=5, learning_rate=1, n_estimators=100
0.7290.041max_features="log2", max_depth=7, learning_rate=1, n_estimators=100
0.6680.112max_features="log2", max_depth=5, learning_rate=1, n_estimators=500
0.630.129max_features="log2", max_depth=5, learning_rate=1, n_estimators=300
0.6080.085max_features="log2", max_depth=7, learning_rate=1, n_estimators=700
0.5960.09max_features="log2", max_depth=7, learning_rate=1, n_estimators=300
0.5860.079max_features="log2", max_depth=5, learning_rate=1, n_estimators=700
0.5360.058max_features="log2", max_depth=7, learning_rate=1, n_estimators=500

BernoulliNB

mean(scores)std(scores)params
---------------:--------------::---------
0.8170.02

RandomForestClassifier

mean(scores)std(scores)params
---------------:--------------::--------------------------------------------------------------------------------
0.90.017min_samples_leaf=13, max_features="log2", n_estimators=640, criterion="entropy"
0.8990.015min_samples_leaf=13, max_features="log2", n_estimators=160, criterion="entropy"
0.8990.016min_samples_leaf=13, max_features="log2", n_estimators=320, criterion="entropy"
0.8990.017min_samples_leaf=5, max_features="log2", n_estimators=640, criterion="entropy"
0.8980.016min_samples_leaf=7, max_features="log2", n_estimators=640, criterion="entropy"
0.8970.016min_samples_leaf=13, max_features="log2", n_estimators=80, criterion="entropy"
0.8970.017min_samples_leaf=5, max_features="log2", n_estimators=320, criterion="entropy"
0.8970.014min_samples_leaf=5, max_features="log2", n_estimators=160, criterion="entropy"
0.8970.018min_samples_leaf=3, max_features="log2", n_estimators=640, criterion="entropy"
0.8960.018min_samples_leaf=7, max_features="log2", n_estimators=320, criterion="entropy"
0.8960.014min_samples_leaf=13, max_features="log2", n_estimators=40, criterion="entropy"
0.8960.014min_samples_leaf=7, max_features="log2", n_estimators=160, criterion="entropy"
0.8960.014min_samples_leaf=13, max_features="log2", n_estimators=320, criterion="gini"
0.8950.013min_samples_leaf=13, max_features="log2", n_estimators=640, criterion="gini"
0.8940.015min_samples_leaf=3, max_features="log2", n_estimators=80, criterion="entropy"
0.8940.016min_samples_leaf=7, max_features="log2", n_estimators=80, criterion="entropy"
0.8930.014min_samples_leaf=7, max_features="log2", n_estimators=640, criterion="gini"
0.8930.014min_samples_leaf=5, max_features="log2", n_estimators=640, criterion="gini"
0.8920.016min_samples_leaf=13, max_features="log2", n_estimators=80, criterion="gini"
0.8920.014min_samples_leaf=7, max_features="log2", n_estimators=160, criterion="gini"
0.8920.018min_samples_leaf=1, max_features="log2", n_estimators=640, criterion="entropy"
0.8920.015min_samples_leaf=3, max_features="log2", n_estimators=160, criterion="entropy"
0.8910.018min_samples_leaf=3, max_features="log2", n_estimators=320, criterion="entropy"
0.8910.015min_samples_leaf=13, max_features="log2", n_estimators=160, criterion="gini"
0.8910.015min_samples_leaf=5, max_features="log2", n_estimators=320, criterion="gini"
0.890.018min_samples_leaf=7, max_features="log2", n_estimators=40, criterion="entropy"
0.890.015min_samples_leaf=7, max_features="log2", n_estimators=320, criterion="gini"
0.890.016min_samples_leaf=5, max_features="log2", n_estimators=80, criterion="entropy"
0.890.015min_samples_leaf=3, max_features="log2", n_estimators=640, criterion="gini"
0.890.014min_samples_leaf=7, max_features="log2", n_estimators=80, criterion="gini"
0.8890.016min_samples_leaf=5, max_features="log2", n_estimators=160, criterion="gini"
0.8890.017min_samples_leaf=1, max_features="log2", n_estimators=160, criterion="entropy"
0.8890.013min_samples_leaf=13, max_features="log2", n_estimators=20, criterion="gini"
0.8890.02min_samples_leaf=13, max_features="log2", n_estimators=20, criterion="entropy"
0.8890.014min_samples_leaf=3, max_features="log2", n_estimators=320, criterion="gini"
0.8880.018min_samples_leaf=3, max_features="log2", n_estimators=40, criterion="entropy"
0.8880.015min_samples_leaf=7, max_features="log2", n_estimators=40, criterion="gini"
0.8880.016min_samples_leaf=1, max_features="log2", n_estimators=320, criterion="entropy"
0.8870.015min_samples_leaf=3, max_features="log2", n_estimators=160, criterion="gini"
0.8870.016min_samples_leaf=1, max_features="log2", n_estimators=640, criterion="gini"
0.8870.018min_samples_leaf=13, max_features="log2", n_estimators=40, criterion="gini"
0.8860.016min_samples_leaf=5, max_features="log2", n_estimators=40, criterion="entropy"
0.8850.02min_samples_leaf=5, max_features="log2", n_estimators=40, criterion="gini"
0.8850.017min_samples_leaf=3, max_features="log2", n_estimators=80, criterion="gini"
0.8840.019min_samples_leaf=5, max_features="log2", n_estimators=80, criterion="gini"
0.8830.014min_samples_leaf=1, max_features="log2", n_estimators=320, criterion="gini"
0.8820.017min_samples_leaf=7, max_features="log2", n_estimators=20, criterion="entropy"
0.8810.021min_samples_leaf=5, max_features="log2", n_estimators=20, criterion="entropy"
0.880.019min_samples_leaf=1, max_features="log2", n_estimators=80, criterion="entropy"
0.8760.01min_samples_leaf=3, max_features="log2", n_estimators=40, criterion="gini"
0.8760.013min_samples_leaf=1, max_features="log2", n_estimators=160, criterion="gini"
0.8750.015min_samples_leaf=7, max_features="log2", n_estimators=20, criterion="gini"
0.8740.011min_samples_leaf=13, max_features="log2", n_estimators=10, criterion="entropy"
0.8740.024min_samples_leaf=13, max_features="log2", n_estimators=10, criterion="gini"
0.870.017min_samples_leaf=5, max_features="log2", n_estimators=20, criterion="gini"
0.8670.017min_samples_leaf=3, max_features="log2", n_estimators=20, criterion="entropy"
0.8650.013min_samples_leaf=1, max_features="log2", n_estimators=80, criterion="gini"
0.8640.015min_samples_leaf=7, max_features="log2", n_estimators=10, criterion="gini"
0.8580.012min_samples_leaf=1, max_features="log2", n_estimators=40, criterion="entropy"
0.8580.014min_samples_leaf=3, max_features="log2", n_estimators=20, criterion="gini"
0.8570.024min_samples_leaf=7, max_features="log2", n_estimators=10, criterion="entropy"
0.8530.024min_samples_leaf=5, max_features="log2", n_estimators=10, criterion="entropy"
0.8520.015min_samples_leaf=5, max_features="log2", n_estimators=10, criterion="gini"
0.850.014min_samples_leaf=1, max_features="log2", n_estimators=40, criterion="gini"
0.8460.025min_samples_leaf=3, max_features="log2", n_estimators=10, criterion="gini"
0.8390.006min_samples_leaf=3, max_features="log2", n_estimators=10, criterion="entropy"
0.8360.013min_samples_leaf=1, max_features="log2", n_estimators=20, criterion="entropy"
0.8190.019min_samples_leaf=1, max_features="log2", n_estimators=20, criterion="gini"
0.7890.004min_samples_leaf=1, max_features="log2", n_estimators=10, criterion="gini"
0.7830.009min_samples_leaf=1, max_features="log2", n_estimators=10, criterion="entropy"

GaussianNB

mean(scores)std(scores)params
---------------:--------------::---------
0.7980.032

After taking Adam's suggestions into account and scoring the sentiment of edits, I still wasn't able to get a decent signal. The 1% rise in accuracy is due to something else, and I think sentiment on edits is not a very good feature. To validate my conclusion further, I took 1000 samples from enwiki, half damaging and half not damaging and here's the result:

Enwiki_1000_damaging_notdamaging_edits_sentiment (480×640 px, 17 KB)

My reasoning for the above result goes like this:
Most of the newer edits on Wikipedia are small or those done with a damaging intent contain random or bad words. Both of these do not fall in the definition of intellectual negative words which might degrade the overall quality of article in a meaningful sense. Hence the sentiment score both positive and negative seems roughly the same for every edit either damaging or not damaging