Microsoft Windows [Version 10.0.16299.904] (c) 2017 Microsoft Corporation. All rights reserved. E:\!CORPORA\CxG-Background-Corpus\!Frontiers>python classify_unmask.py Starting unigrams and twitter (28000, 30000) 28000 {'C': 0.01, 'loss': 'squared_hinge'} STARTING UNMASKING ROUND 1 eng twitter unigrams 1 precision recall f1-score support au 1.00 1.00 1.00 5000 ca 1.00 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 1.00 1.00 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 1.00 1.00 1.00 5000 us 1.00 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 30000) (64208, 30000) (303724, 29979) (64208, 29979) STARTING UNMASKING ROUND 2 eng twitter unigrams 2 precision recall f1-score support au 1.00 1.00 1.00 5000 ca 1.00 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 1.00 1.00 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 1.00 1.00 1.00 5000 us 1.00 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29979) (64208, 29979) (303724, 29957) (64208, 29957) STARTING UNMASKING ROUND 3 eng twitter unigrams 3 precision recall f1-score support au 1.00 1.00 1.00 5000 ca 0.99 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 1.00 1.00 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 1.00 5000 us 1.00 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29957) (64208, 29957) (303724, 29935) (64208, 29935) STARTING UNMASKING ROUND 4 eng twitter unigrams 4 precision recall f1-score support au 1.00 1.00 1.00 5000 ca 1.00 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 1.00 1.00 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 0.99 5000 us 0.99 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29935) (64208, 29935) (303724, 29910) (64208, 29910) STARTING UNMASKING ROUND 5 eng twitter unigrams 5 precision recall f1-score support au 0.99 1.00 1.00 5000 ca 0.99 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 0.99 1.00 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 0.99 5000 us 0.99 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29910) (64208, 29910) (303724, 29884) (64208, 29884) STARTING UNMASKING ROUND 6 eng twitter unigrams 6 precision recall f1-score support au 0.99 1.00 1.00 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 0.99 5000 us 0.99 1.00 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29884) (64208, 29884) (303724, 29857) (64208, 29857) STARTING UNMASKING ROUND 7 eng twitter unigrams 7 precision recall f1-score support au 0.99 1.00 1.00 5000 ca 0.99 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 1.00 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 0.99 5000 us 0.99 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29857) (64208, 29857) (303724, 29832) (64208, 29832) STARTING UNMASKING ROUND 8 eng twitter unigrams 8 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.99 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 1.00 1.00 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29832) (64208, 29832) (303724, 29808) (64208, 29808) STARTING UNMASKING ROUND 9 eng twitter unigrams 9 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 1.00 0.99 5000 us 0.99 1.00 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29808) (64208, 29808) (303724, 29783) (64208, 29783) STARTING UNMASKING ROUND 10 eng twitter unigrams 10 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 1.00 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29783) (64208, 29783) (303724, 29757) (64208, 29757) STARTING UNMASKING ROUND 11 eng twitter unigrams 11 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 1.00 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29757) (64208, 29757) (303724, 29731) (64208, 29731) STARTING UNMASKING ROUND 12 eng twitter unigrams 12 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 1.00 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29731) (64208, 29731) (303724, 29704) (64208, 29704) STARTING UNMASKING ROUND 13 eng twitter unigrams 13 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.99 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29704) (64208, 29704) (303724, 29678) (64208, 29678) STARTING UNMASKING ROUND 14 eng twitter unigrams 14 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.98 5000 ch 1.00 1.00 1.00 3154 ie 1.00 1.00 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29678) (64208, 29678) (303724, 29650) (64208, 29650) STARTING UNMASKING ROUND 15 eng twitter unigrams 15 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29650) (64208, 29650) (303724, 29622) (64208, 29622) STARTING UNMASKING ROUND 16 eng twitter unigrams 16 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.98 0.98 5000 ch 1.00 1.00 1.00 3154 ie 1.00 0.99 1.00 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 1.00 1.00 1.00 64208 Reducing feature vectors. (303724, 29622) (64208, 29622) (303724, 29594) (64208, 29594) STARTING UNMASKING ROUND 17 eng twitter unigrams 17 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29594) (64208, 29594) (303724, 29567) (64208, 29567) STARTING UNMASKING ROUND 18 eng twitter unigrams 18 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29567) (64208, 29567) (303724, 29543) (64208, 29543) STARTING UNMASKING ROUND 19 eng twitter unigrams 19 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29543) (64208, 29543) (303724, 29516) (64208, 29516) STARTING UNMASKING ROUND 20 eng twitter unigrams 20 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29516) (64208, 29516) (303724, 29489) (64208, 29489) STARTING UNMASKING ROUND 21 eng twitter unigrams 21 precision recall f1-score support au 0.99 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29489) (64208, 29489) (303724, 29463) (64208, 29463) STARTING UNMASKING ROUND 22 eng twitter unigrams 22 precision recall f1-score support au 0.98 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29463) (64208, 29463) (303724, 29435) (64208, 29435) STARTING UNMASKING ROUND 23 eng twitter unigrams 23 precision recall f1-score support au 0.98 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29435) (64208, 29435) (303724, 29407) (64208, 29407) STARTING UNMASKING ROUND 24 eng twitter unigrams 24 precision recall f1-score support au 0.98 0.99 0.99 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29407) (64208, 29407) (303724, 29380) (64208, 29380) STARTING UNMASKING ROUND 25 eng twitter unigrams 25 precision recall f1-score support au 0.98 0.99 0.98 5000 ca 0.99 0.97 0.98 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29380) (64208, 29380) (303724, 29352) (64208, 29352) STARTING UNMASKING ROUND 26 eng twitter unigrams 26 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.97 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29352) (64208, 29352) (303724, 29324) (64208, 29324) STARTING UNMASKING ROUND 27 eng twitter unigrams 27 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.97 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.99 0.99 0.99 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29324) (64208, 29324) (303724, 29297) (64208, 29297) STARTING UNMASKING ROUND 28 eng twitter unigrams 28 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.97 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.99 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.99 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29297) (64208, 29297) (303724, 29270) (64208, 29270) STARTING UNMASKING ROUND 29 eng twitter unigrams 29 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29270) (64208, 29270) (303724, 29242) (64208, 29242) STARTING UNMASKING ROUND 30 eng twitter unigrams 30 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.99 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.99 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29242) (64208, 29242) (303724, 29215) (64208, 29215) STARTING UNMASKING ROUND 31 eng twitter unigrams 31 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29215) (64208, 29215) (303724, 29188) (64208, 29188) STARTING UNMASKING ROUND 32 eng twitter unigrams 32 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29188) (64208, 29188) (303724, 29161) (64208, 29161) STARTING UNMASKING ROUND 33 eng twitter unigrams 33 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29161) (64208, 29161) (303724, 29134) (64208, 29134) STARTING UNMASKING ROUND 34 eng twitter unigrams 34 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29134) (64208, 29134) (303724, 29106) (64208, 29106) STARTING UNMASKING ROUND 35 eng twitter unigrams 35 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.99 0.98 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29106) (64208, 29106) (303724, 29079) (64208, 29079) STARTING UNMASKING ROUND 36 eng twitter unigrams 36 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29079) (64208, 29079) (303724, 29051) (64208, 29051) STARTING UNMASKING ROUND 37 eng twitter unigrams 37 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29051) (64208, 29051) (303724, 29023) (64208, 29023) STARTING UNMASKING ROUND 38 eng twitter unigrams 38 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.99 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 29023) (64208, 29023) (303724, 28997) (64208, 28997) STARTING UNMASKING ROUND 39 eng twitter unigrams 39 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.98 0.96 0.97 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.99 0.98 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28997) (64208, 28997) (303724, 28969) (64208, 28969) STARTING UNMASKING ROUND 40 eng twitter unigrams 40 precision recall f1-score support au 0.98 0.98 0.98 5000 ca 0.97 0.95 0.96 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.99 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28969) (64208, 28969) (303724, 28941) (64208, 28941) STARTING UNMASKING ROUND 41 eng twitter unigrams 41 precision recall f1-score support au 0.97 0.98 0.98 5000 ca 0.97 0.95 0.96 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28941) (64208, 28941) (303724, 28914) (64208, 28914) STARTING UNMASKING ROUND 42 eng twitter unigrams 42 precision recall f1-score support au 0.97 0.98 0.98 5000 ca 0.97 0.95 0.96 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.99 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28914) (64208, 28914) (303724, 28886) (64208, 28886) STARTING UNMASKING ROUND 43 eng twitter unigrams 43 precision recall f1-score support au 0.97 0.98 0.97 5000 ca 0.97 0.96 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 1.00 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.99 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28886) (64208, 28886) (303724, 28858) (64208, 28858) STARTING UNMASKING ROUND 44 eng twitter unigrams 44 precision recall f1-score support au 0.97 0.98 0.97 5000 ca 0.97 0.95 0.96 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28858) (64208, 28858) (303724, 28832) (64208, 28832) STARTING UNMASKING ROUND 45 eng twitter unigrams 45 precision recall f1-score support au 0.97 0.98 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28832) (64208, 28832) (303724, 28804) (64208, 28804) STARTING UNMASKING ROUND 46 eng twitter unigrams 46 precision recall f1-score support au 0.97 0.98 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28804) (64208, 28804) (303724, 28776) (64208, 28776) STARTING UNMASKING ROUND 47 eng twitter unigrams 47 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28776) (64208, 28776) (303724, 28748) (64208, 28748) STARTING UNMASKING ROUND 48 eng twitter unigrams 48 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28748) (64208, 28748) (303724, 28720) (64208, 28720) STARTING UNMASKING ROUND 49 eng twitter unigrams 49 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28720) (64208, 28720) (303724, 28692) (64208, 28692) STARTING UNMASKING ROUND 50 eng twitter unigrams 50 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28692) (64208, 28692) (303724, 28665) (64208, 28665) STARTING UNMASKING ROUND 51 eng twitter unigrams 51 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28665) (64208, 28665) (303724, 28637) (64208, 28637) STARTING UNMASKING ROUND 52 eng twitter unigrams 52 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.98 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28637) (64208, 28637) (303724, 28610) (64208, 28610) STARTING UNMASKING ROUND 53 eng twitter unigrams 53 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.95 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28610) (64208, 28610) (303724, 28582) (64208, 28582) STARTING UNMASKING ROUND 54 eng twitter unigrams 54 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.96 5000 ch 0.99 1.00 0.99 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28582) (64208, 28582) (303724, 28554) (64208, 28554) STARTING UNMASKING ROUND 55 eng twitter unigrams 55 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.96 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28554) (64208, 28554) (303724, 28526) (64208, 28526) STARTING UNMASKING ROUND 56 eng twitter unigrams 56 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 1.00 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28526) (64208, 28526) (303724, 28499) (64208, 28499) STARTING UNMASKING ROUND 57 eng twitter unigrams 57 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28499) (64208, 28499) (303724, 28471) (64208, 28471) STARTING UNMASKING ROUND 58 eng twitter unigrams 58 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.98 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28471) (64208, 28471) (303724, 28443) (64208, 28443) STARTING UNMASKING ROUND 59 eng twitter unigrams 59 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28443) (64208, 28443) (303724, 28416) (64208, 28416) STARTING UNMASKING ROUND 60 eng twitter unigrams 60 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28416) (64208, 28416) (303724, 28388) (64208, 28388) STARTING UNMASKING ROUND 61 eng twitter unigrams 61 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28388) (64208, 28388) (303724, 28360) (64208, 28360) STARTING UNMASKING ROUND 62 eng twitter unigrams 62 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.98 0.99 0.98 5000 za 0.99 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28360) (64208, 28360) (303724, 28332) (64208, 28332) STARTING UNMASKING ROUND 63 eng twitter unigrams 63 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28332) (64208, 28332) (303724, 28304) (64208, 28304) STARTING UNMASKING ROUND 64 eng twitter unigrams 64 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.94 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 1.00 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28304) (64208, 28304) (303724, 28276) (64208, 28276) STARTING UNMASKING ROUND 65 eng twitter unigrams 65 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28276) (64208, 28276) (303724, 28249) (64208, 28249) STARTING UNMASKING ROUND 66 eng twitter unigrams 66 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.94 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 0.99 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28249) (64208, 28249) (303724, 28221) (64208, 28221) STARTING UNMASKING ROUND 67 eng twitter unigrams 67 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.98 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28221) (64208, 28221) (303724, 28193) (64208, 28193) STARTING UNMASKING ROUND 68 eng twitter unigrams 68 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.97 0.94 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 1.00 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28193) (64208, 28193) (303724, 28165) (64208, 28165) STARTING UNMASKING ROUND 69 eng twitter unigrams 69 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28165) (64208, 28165) (303724, 28138) (64208, 28138) STARTING UNMASKING ROUND 70 eng twitter unigrams 70 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28138) (64208, 28138) (303724, 28110) (64208, 28110) STARTING UNMASKING ROUND 71 eng twitter unigrams 71 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 1.00 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.99 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.99 0.99 0.99 64208 Reducing feature vectors. (303724, 28110) (64208, 28110) (303724, 28082) (64208, 28082) STARTING UNMASKING ROUND 72 eng twitter unigrams 72 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.98 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 28082) (64208, 28082) (303724, 28054) (64208, 28054) STARTING UNMASKING ROUND 73 eng twitter unigrams 73 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 28054) (64208, 28054) (303724, 28026) (64208, 28026) STARTING UNMASKING ROUND 74 eng twitter unigrams 74 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 1.00 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 28026) (64208, 28026) (303724, 27999) (64208, 27999) STARTING UNMASKING ROUND 75 eng twitter unigrams 75 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27999) (64208, 27999) (303724, 27971) (64208, 27971) STARTING UNMASKING ROUND 76 eng twitter unigrams 76 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27971) (64208, 27971) (303724, 27943) (64208, 27943) STARTING UNMASKING ROUND 77 eng twitter unigrams 77 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.95 5000 ch 0.99 1.00 1.00 3154 ie 0.99 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.98 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27943) (64208, 27943) (303724, 27916) (64208, 27916) STARTING UNMASKING ROUND 78 eng twitter unigrams 78 precision recall f1-score support au 0.97 0.97 0.97 5000 ca 0.96 0.93 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27916) (64208, 27916) (303724, 27889) (64208, 27889) STARTING UNMASKING ROUND 79 eng twitter unigrams 79 precision recall f1-score support au 0.96 0.97 0.96 5000 ca 0.96 0.93 0.94 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.99 0.99 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27889) (64208, 27889) (303724, 27861) (64208, 27861) STARTING UNMASKING ROUND 80 eng twitter unigrams 80 precision recall f1-score support au 0.96 0.97 0.97 5000 ca 0.96 0.93 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27861) (64208, 27861) (303724, 27833) (64208, 27833) STARTING UNMASKING ROUND 81 eng twitter unigrams 81 precision recall f1-score support au 0.96 0.97 0.96 5000 ca 0.96 0.93 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27833) (64208, 27833) (303724, 27805) (64208, 27805) STARTING UNMASKING ROUND 82 eng twitter unigrams 82 precision recall f1-score support au 0.96 0.97 0.96 5000 ca 0.96 0.93 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.97 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27805) (64208, 27805) (303724, 27778) (64208, 27778) STARTING UNMASKING ROUND 83 eng twitter unigrams 83 precision recall f1-score support au 0.96 0.97 0.96 5000 ca 0.96 0.93 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27778) (64208, 27778) (303724, 27750) (64208, 27750) STARTING UNMASKING ROUND 84 eng twitter unigrams 84 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.93 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.97 0.98 0.97 5000 za 1.00 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27750) (64208, 27750) (303724, 27723) (64208, 27723) STARTING UNMASKING ROUND 85 eng twitter unigrams 85 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.93 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.99 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.98 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27723) (64208, 27723) (303724, 27695) (64208, 27695) STARTING UNMASKING ROUND 86 eng twitter unigrams 86 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.93 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.97 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27695) (64208, 27695) (303724, 27667) (64208, 27667) STARTING UNMASKING ROUND 87 eng twitter unigrams 87 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.93 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.97 0.97 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27667) (64208, 27667) (303724, 27639) (64208, 27639) STARTING UNMASKING ROUND 88 eng twitter unigrams 88 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.92 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.97 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27639) (64208, 27639) (303724, 27611) (64208, 27611) STARTING UNMASKING ROUND 89 eng twitter unigrams 89 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.92 0.94 5000 ch 1.00 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27611) (64208, 27611) (303724, 27583) (64208, 27583) STARTING UNMASKING ROUND 90 eng twitter unigrams 90 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27583) (64208, 27583) (303724, 27555) (64208, 27555) STARTING UNMASKING ROUND 91 eng twitter unigrams 91 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27555) (64208, 27555) (303724, 27527) (64208, 27527) STARTING UNMASKING ROUND 92 eng twitter unigrams 92 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.96 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27527) (64208, 27527) (303724, 27499) (64208, 27499) STARTING UNMASKING ROUND 93 eng twitter unigrams 93 precision recall f1-score support au 0.96 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.98 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27499) (64208, 27499) (303724, 27472) (64208, 27472) STARTING UNMASKING ROUND 94 eng twitter unigrams 94 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.98 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27472) (64208, 27472) (303724, 27444) (64208, 27444) STARTING UNMASKING ROUND 95 eng twitter unigrams 95 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27444) (64208, 27444) (303724, 27416) (64208, 27416) STARTING UNMASKING ROUND 96 eng twitter unigrams 96 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27416) (64208, 27416) (303724, 27388) (64208, 27388) STARTING UNMASKING ROUND 97 eng twitter unigrams 97 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27388) (64208, 27388) (303724, 27360) (64208, 27360) STARTING UNMASKING ROUND 98 eng twitter unigrams 98 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.94 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 1.00 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27360) (64208, 27360) (303724, 27333) (64208, 27333) STARTING UNMASKING ROUND 99 eng twitter unigrams 99 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27333) (64208, 27333) (303724, 27306) (64208, 27306) STARTING UNMASKING ROUND 100 eng twitter unigrams 100 precision recall f1-score support au 0.95 0.96 0.96 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27306) (64208, 27306) (303724, 27278) (64208, 27278) STARTING UNMASKING ROUND 101 eng twitter unigrams 101 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27278) (64208, 27278) (303724, 27251) (64208, 27251) STARTING UNMASKING ROUND 102 eng twitter unigrams 102 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27251) (64208, 27251) (303724, 27223) (64208, 27223) STARTING UNMASKING ROUND 103 eng twitter unigrams 103 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27223) (64208, 27223) (303724, 27195) (64208, 27195) STARTING UNMASKING ROUND 104 eng twitter unigrams 104 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27195) (64208, 27195) (303724, 27168) (64208, 27168) STARTING UNMASKING ROUND 105 eng twitter unigrams 105 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 1.00 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27168) (64208, 27168) (303724, 27140) (64208, 27140) STARTING UNMASKING ROUND 106 eng twitter unigrams 106 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27140) (64208, 27140) (303724, 27112) (64208, 27112) STARTING UNMASKING ROUND 107 eng twitter unigrams 107 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27112) (64208, 27112) (303724, 27084) (64208, 27084) STARTING UNMASKING ROUND 108 eng twitter unigrams 108 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27084) (64208, 27084) (303724, 27056) (64208, 27056) STARTING UNMASKING ROUND 109 eng twitter unigrams 109 precision recall f1-score support au 0.95 0.96 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27056) (64208, 27056) (303724, 27029) (64208, 27029) STARTING UNMASKING ROUND 110 eng twitter unigrams 110 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27029) (64208, 27029) (303724, 27002) (64208, 27002) STARTING UNMASKING ROUND 111 eng twitter unigrams 111 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 27002) (64208, 27002) (303724, 26974) (64208, 26974) STARTING UNMASKING ROUND 112 eng twitter unigrams 112 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26974) (64208, 26974) (303724, 26946) (64208, 26946) STARTING UNMASKING ROUND 113 eng twitter unigrams 113 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.92 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26946) (64208, 26946) (303724, 26918) (64208, 26918) STARTING UNMASKING ROUND 114 eng twitter unigrams 114 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.98 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26918) (64208, 26918) (303724, 26891) (64208, 26891) STARTING UNMASKING ROUND 115 eng twitter unigrams 115 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26891) (64208, 26891) (303724, 26863) (64208, 26863) STARTING UNMASKING ROUND 116 eng twitter unigrams 116 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26863) (64208, 26863) (303724, 26835) (64208, 26835) STARTING UNMASKING ROUND 117 eng twitter unigrams 117 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26835) (64208, 26835) (303724, 26807) (64208, 26807) STARTING UNMASKING ROUND 118 eng twitter unigrams 118 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 0.99 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26807) (64208, 26807) (303724, 26779) (64208, 26779) STARTING UNMASKING ROUND 119 eng twitter unigrams 119 precision recall f1-score support au 0.95 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 0.99 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.97 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26779) (64208, 26779) (303724, 26751) (64208, 26751) STARTING UNMASKING ROUND 120 eng twitter unigrams 120 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26751) (64208, 26751) (303724, 26723) (64208, 26723) STARTING UNMASKING ROUND 121 eng twitter unigrams 121 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 0.99 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26723) (64208, 26723) (303724, 26695) (64208, 26695) STARTING UNMASKING ROUND 122 eng twitter unigrams 122 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26695) (64208, 26695) (303724, 26667) (64208, 26667) STARTING UNMASKING ROUND 123 eng twitter unigrams 123 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26667) (64208, 26667) (303724, 26639) (64208, 26639) STARTING UNMASKING ROUND 124 eng twitter unigrams 124 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26639) (64208, 26639) (303724, 26611) (64208, 26611) STARTING UNMASKING ROUND 125 eng twitter unigrams 125 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.95 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26611) (64208, 26611) (303724, 26583) (64208, 26583) STARTING UNMASKING ROUND 126 eng twitter unigrams 126 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.97 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26583) (64208, 26583) (303724, 26555) (64208, 26555) STARTING UNMASKING ROUND 127 eng twitter unigrams 127 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26555) (64208, 26555) (303724, 26527) (64208, 26527) STARTING UNMASKING ROUND 128 eng twitter unigrams 128 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.91 0.93 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26527) (64208, 26527) (303724, 26501) (64208, 26501) STARTING UNMASKING ROUND 129 eng twitter unigrams 129 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26501) (64208, 26501) (303724, 26475) (64208, 26475) STARTING UNMASKING ROUND 130 eng twitter unigrams 130 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26475) (64208, 26475) (303724, 26447) (64208, 26447) STARTING UNMASKING ROUND 131 eng twitter unigrams 131 precision recall f1-score support au 0.94 0.95 0.95 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26447) (64208, 26447) (303724, 26419) (64208, 26419) STARTING UNMASKING ROUND 132 eng twitter unigrams 132 precision recall f1-score support au 0.94 0.95 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26419) (64208, 26419) (303724, 26391) (64208, 26391) STARTING UNMASKING ROUND 133 eng twitter unigrams 133 precision recall f1-score support au 0.94 0.95 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.97 5000 us 0.96 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26391) (64208, 26391) (303724, 26363) (64208, 26363) STARTING UNMASKING ROUND 134 eng twitter unigrams 134 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26363) (64208, 26363) (303724, 26335) (64208, 26335) STARTING UNMASKING ROUND 135 eng twitter unigrams 135 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.96 0.96 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26335) (64208, 26335) (303724, 26307) (64208, 26307) STARTING UNMASKING ROUND 136 eng twitter unigrams 136 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26307) (64208, 26307) (303724, 26279) (64208, 26279) STARTING UNMASKING ROUND 137 eng twitter unigrams 137 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 1.00 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26279) (64208, 26279) (303724, 26251) (64208, 26251) STARTING UNMASKING ROUND 138 eng twitter unigrams 138 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26251) (64208, 26251) (303724, 26223) (64208, 26223) STARTING UNMASKING ROUND 139 eng twitter unigrams 139 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.99 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26223) (64208, 26223) (303724, 26195) (64208, 26195) STARTING UNMASKING ROUND 140 eng twitter unigrams 140 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.91 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.96 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.98 0.98 0.98 64208 Reducing feature vectors. (303724, 26195) (64208, 26195) (303724, 26167) (64208, 26167) STARTING UNMASKING ROUND 141 eng twitter unigrams 141 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26167) (64208, 26167) (303724, 26140) (64208, 26140) STARTING UNMASKING ROUND 142 eng twitter unigrams 142 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26140) (64208, 26140) (303724, 26112) (64208, 26112) STARTING UNMASKING ROUND 143 eng twitter unigrams 143 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26112) (64208, 26112) (303724, 26084) (64208, 26084) STARTING UNMASKING ROUND 144 eng twitter unigrams 144 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.98 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26084) (64208, 26084) (303724, 26057) (64208, 26057) STARTING UNMASKING ROUND 145 eng twitter unigrams 145 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.94 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26057) (64208, 26057) (303724, 26029) (64208, 26029) STARTING UNMASKING ROUND 146 eng twitter unigrams 146 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26029) (64208, 26029) (303724, 26003) (64208, 26003) STARTING UNMASKING ROUND 147 eng twitter unigrams 147 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.97 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 26003) (64208, 26003) (303724, 25975) (64208, 25975) STARTING UNMASKING ROUND 148 eng twitter unigrams 148 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25975) (64208, 25975) (303724, 25948) (64208, 25948) STARTING UNMASKING ROUND 149 eng twitter unigrams 149 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.92 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25948) (64208, 25948) (303724, 25921) (64208, 25921) STARTING UNMASKING ROUND 150 eng twitter unigrams 150 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.91 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25921) (64208, 25921) (303724, 25893) (64208, 25893) STARTING UNMASKING ROUND 151 eng twitter unigrams 151 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.90 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25893) (64208, 25893) (303724, 25866) (64208, 25866) STARTING UNMASKING ROUND 152 eng twitter unigrams 152 precision recall f1-score support au 0.94 0.94 0.94 5000 ca 0.93 0.89 0.91 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25866) (64208, 25866) (303724, 25838) (64208, 25838) STARTING UNMASKING ROUND 153 eng twitter unigrams 153 precision recall f1-score support au 0.94 0.93 0.94 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25838) (64208, 25838) (303724, 25810) (64208, 25810) STARTING UNMASKING ROUND 154 eng twitter unigrams 154 precision recall f1-score support au 0.94 0.93 0.94 5000 ca 0.93 0.90 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25810) (64208, 25810) (303724, 25782) (64208, 25782) STARTING UNMASKING ROUND 155 eng twitter unigrams 155 precision recall f1-score support au 0.94 0.93 0.93 5000 ca 0.93 0.90 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25782) (64208, 25782) (303724, 25755) (64208, 25755) STARTING UNMASKING ROUND 156 eng twitter unigrams 156 precision recall f1-score support au 0.94 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 1.00 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25755) (64208, 25755) (303724, 25727) (64208, 25727) STARTING UNMASKING ROUND 157 eng twitter unigrams 157 precision recall f1-score support au 0.94 0.93 0.93 5000 ca 0.93 0.90 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.98 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25727) (64208, 25727) (303724, 25699) (64208, 25699) STARTING UNMASKING ROUND 158 eng twitter unigrams 158 precision recall f1-score support au 0.94 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.95 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25699) (64208, 25699) (303724, 25671) (64208, 25671) STARTING UNMASKING ROUND 159 eng twitter unigrams 159 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25671) (64208, 25671) (303724, 25643) (64208, 25643) STARTING UNMASKING ROUND 160 eng twitter unigrams 160 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.96 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25643) (64208, 25643) (303724, 25615) (64208, 25615) STARTING UNMASKING ROUND 161 eng twitter unigrams 161 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25615) (64208, 25615) (303724, 25587) (64208, 25587) STARTING UNMASKING ROUND 162 eng twitter unigrams 162 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.95 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25587) (64208, 25587) (303724, 25559) (64208, 25559) STARTING UNMASKING ROUND 163 eng twitter unigrams 163 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25559) (64208, 25559) (303724, 25531) (64208, 25531) STARTING UNMASKING ROUND 164 eng twitter unigrams 164 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.95 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25531) (64208, 25531) (303724, 25504) (64208, 25504) STARTING UNMASKING ROUND 165 eng twitter unigrams 165 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25504) (64208, 25504) (303724, 25476) (64208, 25476) STARTING UNMASKING ROUND 166 eng twitter unigrams 166 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.93 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.97 0.96 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25476) (64208, 25476) (303724, 25448) (64208, 25448) STARTING UNMASKING ROUND 167 eng twitter unigrams 167 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.98 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.97 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25448) (64208, 25448) (303724, 25420) (64208, 25420) STARTING UNMASKING ROUND 168 eng twitter unigrams 168 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.97 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25420) (64208, 25420) (303724, 25393) (64208, 25393) STARTING UNMASKING ROUND 169 eng twitter unigrams 169 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.95 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25393) (64208, 25393) (303724, 25365) (64208, 25365) STARTING UNMASKING ROUND 170 eng twitter unigrams 170 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25365) (64208, 25365) (303724, 25337) (64208, 25337) STARTING UNMASKING ROUND 171 eng twitter unigrams 171 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.95 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25337) (64208, 25337) (303724, 25309) (64208, 25309) STARTING UNMASKING ROUND 172 eng twitter unigrams 172 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25309) (64208, 25309) (303724, 25281) (64208, 25281) STARTING UNMASKING ROUND 173 eng twitter unigrams 173 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.99 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25281) (64208, 25281) (303724, 25253) (64208, 25253) STARTING UNMASKING ROUND 174 eng twitter unigrams 174 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.91 5000 ch 0.99 1.00 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25253) (64208, 25253) (303724, 25225) (64208, 25225) STARTING UNMASKING ROUND 175 eng twitter unigrams 175 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25225) (64208, 25225) (303724, 25197) (64208, 25197) STARTING UNMASKING ROUND 176 eng twitter unigrams 176 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.89 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.96 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25197) (64208, 25197) (303724, 25169) (64208, 25169) STARTING UNMASKING ROUND 177 eng twitter unigrams 177 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.96 0.96 0.96 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25169) (64208, 25169) (303724, 25141) (64208, 25141) STARTING UNMASKING ROUND 178 eng twitter unigrams 178 precision recall f1-score support au 0.92 0.93 0.93 5000 ca 0.92 0.89 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25141) (64208, 25141) (303724, 25113) (64208, 25113) STARTING UNMASKING ROUND 179 eng twitter unigrams 179 precision recall f1-score support au 0.93 0.93 0.93 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25113) (64208, 25113) (303724, 25085) (64208, 25085) STARTING UNMASKING ROUND 180 eng twitter unigrams 180 precision recall f1-score support au 0.93 0.92 0.93 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25085) (64208, 25085) (303724, 25058) (64208, 25058) STARTING UNMASKING ROUND 181 eng twitter unigrams 181 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25058) (64208, 25058) (303724, 25030) (64208, 25030) STARTING UNMASKING ROUND 182 eng twitter unigrams 182 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25030) (64208, 25030) (303724, 25002) (64208, 25002) STARTING UNMASKING ROUND 183 eng twitter unigrams 183 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 25002) (64208, 25002) (303724, 24974) (64208, 24974) STARTING UNMASKING ROUND 184 eng twitter unigrams 184 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24974) (64208, 24974) (303724, 24946) (64208, 24946) STARTING UNMASKING ROUND 185 eng twitter unigrams 185 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24946) (64208, 24946) (303724, 24918) (64208, 24918) STARTING UNMASKING ROUND 186 eng twitter unigrams 186 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.96 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24918) (64208, 24918) (303724, 24890) (64208, 24890) STARTING UNMASKING ROUND 187 eng twitter unigrams 187 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24890) (64208, 24890) (303724, 24862) (64208, 24862) STARTING UNMASKING ROUND 188 eng twitter unigrams 188 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24862) (64208, 24862) (303724, 24834) (64208, 24834) STARTING UNMASKING ROUND 189 eng twitter unigrams 189 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24834) (64208, 24834) (303724, 24806) (64208, 24806) STARTING UNMASKING ROUND 190 eng twitter unigrams 190 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24806) (64208, 24806) (303724, 24779) (64208, 24779) STARTING UNMASKING ROUND 191 eng twitter unigrams 191 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24779) (64208, 24779) (303724, 24751) (64208, 24751) STARTING UNMASKING ROUND 192 eng twitter unigrams 192 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.98 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24751) (64208, 24751) (303724, 24723) (64208, 24723) STARTING UNMASKING ROUND 193 eng twitter unigrams 193 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.90 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 0.99 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24723) (64208, 24723) (303724, 24695) (64208, 24695) STARTING UNMASKING ROUND 194 eng twitter unigrams 194 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.88 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24695) (64208, 24695) (303724, 24667) (64208, 24667) STARTING UNMASKING ROUND 195 eng twitter unigrams 195 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.94 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.93 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24667) (64208, 24667) (303724, 24639) (64208, 24639) STARTING UNMASKING ROUND 196 eng twitter unigrams 196 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 0.99 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.93 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24639) (64208, 24639) (303724, 24611) (64208, 24611) STARTING UNMASKING ROUND 197 eng twitter unigrams 197 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.92 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 0.99 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.93 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24611) (64208, 24611) (303724, 24583) (64208, 24583) STARTING UNMASKING ROUND 198 eng twitter unigrams 198 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24583) (64208, 24583) (303724, 24555) (64208, 24555) STARTING UNMASKING ROUND 199 eng twitter unigrams 199 precision recall f1-score support au 0.92 0.92 0.92 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.93 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24555) (64208, 24555) (303724, 24527) (64208, 24527) STARTING UNMASKING ROUND 200 eng twitter unigrams 200 precision recall f1-score support au 0.91 0.92 0.91 5000 ca 0.92 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 1.00 0.99 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.93 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.97 0.97 0.97 64208 Reducing feature vectors. (303724, 24527) (64208, 24527) (303724, 24499) (64208, 24499) STARTING UNMASKING ROUND 201 eng twitter unigrams 201 precision recall f1-score support au 0.91 0.92 0.92 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.96 0.97 0.96 64208 Reducing feature vectors. (303724, 24499) (64208, 24499) (303724, 24471) (64208, 24471) STARTING UNMASKING ROUND 202 eng twitter unigrams 202 precision recall f1-score support au 0.91 0.92 0.92 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.99 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 0.99 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.96 0.97 0.96 64208 Reducing feature vectors. (303724, 24471) (64208, 24471) (303724, 24444) (64208, 24444) STARTING UNMASKING ROUND 203 eng twitter unigrams 203 precision recall f1-score support au 0.91 0.92 0.92 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.98 0.99 0.99 5000 nz 0.93 0.94 0.94 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.95 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.96 0.97 0.96 64208 Reducing feature vectors. (303724, 24444) (64208, 24444) (303724, 24416) (64208, 24416) STARTING UNMASKING ROUND 204 eng twitter unigrams 204 precision recall f1-score support au 0.91 0.92 0.91 5000 ca 0.91 0.87 0.89 5000 ch 0.99 0.99 0.99 3154 ie 0.97 0.97 0.97 5000 in 0.99 1.00 1.00 5000 my 1.00 1.00 1.00 3255 ng 0.98 0.99 0.99 5000 nz 0.93 0.94 0.93 5000 ph 1.00 1.00 1.00 3537 pk 1.00 1.00 1.00 5000 pt 0.99 0.98 0.98 4262 uk 0.95 0.95 0.95 5000 us 0.94 0.96 0.95 5000 za 0.99 0.99 0.99 5000 avg / total 0.96 0.96 0.96 64208 Reducing feature vectors. (303724, 24416) (64208, 24416) (303724, 24389) (64208, 24389) STARTING UNMASKING ROUND 205