Research:Revision scoring as a service/Clustering in reverted edits

From Meta, a Wikimedia project coordination wiki

This is result of clustering using K-means algorithm. Code is here. First number is number of reverted edits, second one is coordinate of cluster centers (note that these features are scaled and normalized). The third one is cost function per number of clusters and last one is difference between them.

Note that Turkish Wikipedia and Indonesian Wikipedia have features with zero variance which needs investigation.

eswiki[edit]

[[ -2.13523004e-02  -5.91778441e-02   1.89155569e+00   7.43973904e-02
   -1.23832256e-03   1.60656602e-01   2.02524859e+00   1.32326917e-01
    1.78527632e+00   2.01790502e-01   1.33908272e+00  -6.33535919e-02
   -1.60343628e-01  -1.82951731e-02   8.43537544e-03   6.55124383e-01
    1.15918773e+00   2.25191136e-01   2.10338790e+00   1.93856412e-01
    2.07284512e+00   1.41721855e-01   2.05365954e+00  -1.07589482e+00
   -4.63417110e-02   2.34394340e-02  -1.72448716e-01   4.72070860e-02
    8.91375784e-02  -2.00074680e-01  -5.05721737e-02  -3.58629539e-02
   -2.41270639e-01   1.16438839e-01   1.08165311e+00   1.83662258e-01
    2.02368412e+00  -9.89976923e-02  -2.35326250e-02  -2.17387505e-01
    3.69439946e-01]
 [  3.74545599e-03   1.03805214e-02  -3.31802122e-01  -1.30502169e-02
    2.17217000e-04  -2.81811430e-02  -3.55253501e-01  -2.32117681e-02
   -3.13159415e-01  -3.53965348e-02  -2.34891572e-01   1.11129989e-02
    2.81262436e-02   3.20919828e-03  -1.47966856e-03  -1.14916871e-01
   -2.03335778e-01  -3.95012937e-02  -3.68960096e-01  -3.40047979e-02
   -3.63602518e-01  -2.48597557e-02  -3.60237130e-01   1.88725178e-01
    8.12890583e-03  -4.11156489e-03   3.02496249e-02  -8.28070323e-03
   -1.56358271e-02   3.50955585e-02   8.87098100e-03   6.29080301e-03
    4.23218362e-02  -2.04248039e-02  -1.89735253e-01  -3.22166179e-02
   -3.54979073e-01   1.73654123e-02   4.12791173e-03   3.81324410e-02
   -6.48043084e-02]]
{1: 5.4512249681096216, 2: 5.0087062902460016, 3: 4.8128716005817767, 4: 4.663465268844158, 5: 4.5864753216857022, 6: 4.440538913509319, 7: 4.3422159897994854, 8: 4.2540681569561789, 9: 4.2174133988054017, 10: 4.2235806811133996, 11: 4.0618730495689377}
2 - 1 : -0.442518677864
3 - 2 : -0.195834689664
4 - 3 : -0.149406331738
5 - 4 : -0.0769899471585
6 - 5 : -0.145936408176
7 - 6 : -0.0983229237098
8 - 7 : -0.0881478328433
9 - 8 : -0.0366547581508
10 - 9 : 0.006167282308

fawiki[edit]

699
[[  2.69987333e-03  -7.39019968e-02  -3.18594214e-01  -2.83240533e-02
   -4.70320489e-02  -6.81378578e-02  -2.94888286e-01  -1.06719610e-01
   -3.08139335e-01  -1.86951111e-02  -1.94972986e-01   2.75419239e-03
   -4.91181568e-04   1.48927956e-02  -2.29769180e-02  -2.08356864e-01
   -2.65457256e-01  -8.70065567e-02  -3.36613143e-01  -6.54332399e-02
   -2.38094544e-01  -8.80776707e-02  -3.36929975e-01   6.34561168e-02
    1.09060421e-02  -1.27749804e-02  -7.29011504e-03  -5.30490530e-02
   -1.75698637e-02   1.64309805e-02   4.29342034e-03   1.02697346e-02
   -1.00097678e-01  -3.40675653e-01  -1.62577659e-02  -1.26253172e-01]
 [ -1.52735691e-02   4.18074153e-01   1.80233298e+00   1.60233216e-01
    2.66067020e-01   3.85465596e-01   1.66822516e+00   6.03728082e-01
    1.74318824e+00   1.05760914e-01   1.10299003e+00  -1.55808598e-02
    2.77868430e-03  -8.42506722e-02   1.29983707e-01   1.17870454e+00
    1.50172962e+00   4.92208521e-01   1.90426864e+00   3.70165186e-01
    1.34693485e+00   4.98267965e-01   1.90606100e+00  -3.58980318e-01
   -6.16970383e-02   7.22698894e-02   4.12412222e-02   3.00106071e-01
    9.93952287e-02  -9.29524040e-02  -2.42884922e-02  -5.80973559e-02
    5.66266867e-01   1.92725084e+00   9.19725040e-02   7.14232232e-01]]
{1: 5.2692827826698698, 2: 4.9054959688037814, 3: 4.6341994726108542, 4: 4.4643614940761616, 5: 4.3533284937537351, 6: 4.2651092880745534, 7: 4.14679287858743, 8: 3.9914801800049662, 9: 3.9935538426944661, 10: 3.872458312004321, 11: 3.8050293721986419}
2 - 1 : -0.363786813866
3 - 2 : -0.271296496193
4 - 3 : -0.169837978535
5 - 4 : -0.111033000322
6 - 5 : -0.0882192056792
7 - 6 : -0.118316409487
8 - 7 : -0.155312698582
9 - 8 : 0.0020736626895
10 - 9 : -0.12109553069

enwiki[edit]

1482
[[  6.02661374e-02   6.74978299e-01   1.91954069e+00   7.97930155e-02
    2.75379731e-01   6.93602987e-01   2.07479501e+00   6.79337668e-01
    1.87396848e+00   8.98315193e-01   1.41208975e+00  -1.07766881e-01
   -1.71149168e-01  -7.73345229e-02  -2.34052923e-02   1.18505947e+00
    1.48733692e+00   7.73729688e-01   2.12054762e+00   8.30078964e-01
    2.12501993e+00   8.56702909e-01   2.12411750e+00   6.43110695e-02
    2.22510173e-01  -2.09563606e-01  -1.13144290e-01  -1.30655143e-01
    1.79703191e-01  -1.70171226e-01   2.20809249e-01   5.53887171e-02
    1.19354939e-01   5.98078387e-01   1.02052199e+00   9.22878264e-01
    2.06732055e+00  -7.21061348e-02   1.56160814e-01  -6.47287617e-03
    2.11332107e-01]
 [ -7.96489059e-03  -8.92064521e-02  -2.53690252e-01  -1.05456010e-02
   -3.63947238e-02  -9.16679273e-02  -2.74208966e-01  -8.97825948e-02
   -2.47667339e-01  -1.18723093e-01  -1.86624543e-01   1.42426817e-02
    2.26194087e-02   1.02206818e-02   3.09328921e-03  -1.56619778e-01
   -1.96569357e-01  -1.02257629e-01  -2.80255720e-01  -1.09704859e-01
   -2.80846790e-01  -1.13223532e-01  -2.80727522e-01  -8.49947672e-03
   -2.94073796e-02   2.76963361e-02   1.49533706e-02   1.72676393e-02
   -2.37499251e-02   2.24901620e-02  -2.91825821e-02  -7.32028118e-03
   -1.57741822e-02  -7.90432093e-02  -1.34874181e-01  -1.21969396e-01
   -2.73221126e-01   9.52968779e-03  -2.06385185e-02   8.55467973e-04
   -2.79300646e-02]]
{1: 5.4015620241587161, 2: 5.0577365748516883, 3: 4.8366661833292444, 4: 4.668603460438038, 5: 4.5598336485051174, 6: 4.4728693873162992, 7: 4.2994706784188894, 8: 4.2600435679462789, 9: 4.0895815678292085, 10: 4.095353930029467, 11: 4.1007806792764159}
2 - 1 : -0.343825449307
3 - 2 : -0.221070391522
4 - 3 : -0.168062722891
5 - 4 : -0.108769811933
6 - 5 : -0.0869642611888
7 - 6 : -0.173398708897
8 - 7 : -0.0394271104726
9 - 8 : -0.170462000117
10 - 9 : 0.00577236220026

frwiki[edit]

834
[[ 0.02497493 -0.01132827 -0.30430984  0.00806259 -0.03550952 -0.0329419
  -0.32993716 -0.03889011 -0.30902996 -0.07409582 -0.22842587  0.02185577
   0.02263259  0.03007904  0.01009644 -0.11423691 -0.20163812 -0.02849334
  -0.32853945 -0.02921974 -0.33041927 -0.03547682 -0.3297085   0.11480327
   0.00539598 -0.00360733  0.023732   -0.02209277  0.01004428 -0.00683607
   0.02124895 -0.02242938  0.01013527 -0.03912799 -0.25941914 -0.0508376
  -0.32972616  0.00401762 -0.04973143  0.00910693 -0.05747562]
 [-0.16438046  0.0745606   2.00291205 -0.0530665   0.2337172   0.21681759
   2.17158641  0.25596767  2.03397903  0.48768522  1.50345753 -0.14385067
  -0.14896362 -0.19797479 -0.06645294  0.75188656  1.32714546  0.18753799
   2.16238693  0.19231904  2.17475953  0.23350198  2.1700814  -0.75561424
  -0.03551536  0.02374281 -0.15619969  0.14541063 -0.06610959  0.04499379
  -0.13985672  0.14762613 -0.06670853  0.2575333   1.70744964  0.33460385
   2.17019766 -0.02644324  0.32732325 -0.05994018  0.37829411]]
{1: 5.5540838927851723, 2: 5.1342107431086816, 3: 4.8653650537575341, 4: 4.6904242212884446, 5: 4.5875144956681417, 6: 4.5661905629272201, 7: 4.4198060979381699, 8: 4.3597142707826348, 9: 4.2065391232204075, 10: 4.1600170668273311, 11: 4.0919809937409379}
2 - 1 : -0.419873149676
3 - 2 : -0.268845689351
4 - 3 : -0.174940832469
5 - 4 : -0.10290972562
6 - 5 : -0.0213239327409
7 - 6 : -0.146384464989
8 - 7 : -0.0600918271555
9 - 8 : -0.153175147562
10 - 9 : -0.0465220563931

ptwiki[edit]

1890
[[  1.02230134e-01   7.65861822e-01   1.62462492e+00   2.40610546e-01
    1.16721161e-01   7.63308223e-01   1.63746803e+00   7.34097641e-01
    1.53846584e+00   6.08123184e-01   1.20058390e+00  -7.17954632e-03
   -1.45712955e-01   4.46432744e-02  -2.24007004e-02   1.17109214e+00
    1.39133231e+00   8.68010201e-01   1.76617532e+00   8.46035287e-01
    1.70846976e+00   8.44967644e-01   1.70813771e+00  -6.10483389e-02
    1.63805903e-01  -1.26000517e-01  -1.78778480e-01  -1.63444481e-02
    1.52004371e-01  -1.01976044e-01   8.63073694e-02   1.83245857e-01
    1.32712553e-01   6.70739539e-01   1.28432911e+00   9.37445166e-01
    1.69886693e+00  -6.76811705e-04   8.99045702e-02   1.46875443e-01
    5.03012224e-01]
 [ -1.96720446e-02  -1.47374041e-01  -3.12624984e-01  -4.63004520e-02
   -2.24605389e-02  -1.46882655e-01  -3.15096372e-01  -1.41261691e-01
   -2.96045478e-01  -1.17020550e-01  -2.31027185e-01   1.38155308e-03
    2.80394014e-02  -8.59066164e-03   4.31054488e-03  -2.25352115e-01
   -2.67732716e-01  -1.67030354e-01  -3.39863390e-01  -1.62801743e-01
   -3.28759165e-01  -1.62596298e-01  -3.28695268e-01   1.17474721e-02
   -3.15210098e-02   2.42461562e-02   3.44021680e-02   3.14514616e-03
   -2.92500525e-02   1.96231504e-02  -1.66080427e-02  -3.52618210e-02
   -2.55377469e-02  -1.29069753e-01  -2.47142195e-01  -1.80391657e-01
   -3.26911303e-01   1.30238215e-04  -1.73002485e-02  -2.82630979e-02
   -9.67941504e-02]]
{1: 5.5988156010161925, 2: 5.1708824726595557, 3: 4.9168963251299438, 4: 4.7942964356990894, 5: 4.6859694678632682, 6: 4.5505909303082808, 7: 4.5554082320940097, 8: 4.3567669110909604, 9: 4.3013034556054581, 10: 4.3269366395296549, 11: 4.1597569939088404}
2 - 1 : -0.427933128357
3 - 2 : -0.25398614753
4 - 3 : -0.122599889431
5 - 4 : -0.108326967836
6 - 5 : -0.135378537555
7 - 6 : 0.00481730178573
8 - 7 : -0.198641321003
9 - 8 : -0.0554634554855
10 - 9 : 0.0256331839242