P13-1117 results . Table 4 presents the unsupervised evaluation re - sults . Note that the unsupervised
S07-1002 gold standard . Table 3 shows the unsupervised evaluation of the systems on the test corpus
P14-2131 end , we use a metric based on unsupervised evaluation of POS tag - gers . We perform
S07-1075 Table 1 shows the results of the unsupervised evaluation in task 2 , where our system
S07-1002 entropy and purity measures in the unsupervised evaluation . The entropy measure considers
S07-1087 points ( 81.6 - 75.6 ) . In the unsupervised evaluation of this same data this difference
P11-1148 are averaged . 4.4 Results 4.4.1 Unsupervised evaluation In table 1 , we present the performance
S10-1011 the results are averaged . 4.1 Unsupervised evaluation Table 4 shows the V-Measure (
S07-1075 evaluation measures are provided : and unsupervised evaluation ( FScore measure ) and a supervised
S07-1087 3.1 Unsupervised Evaluation The unsupervised evaluation was based on the traditional
E06-1018 novel and likewise automatic and unsupervised evaluation method inspired by Sch " utze
N07-3010 Bordag ( 2006 ) , tested using the unsupervised evaluation framework presented there . More
S07-1002 information to the more standard unsupervised evaluation . In our case , we noticed that
D13-1148 systems and discuss two metrics for unsupervised evaluation ( V - Measure , paired F-Score
S07-1087 the SenseClusters method . 3.1 Unsupervised Evaluation The unsupervised evaluation was
S07-1002 see each of them in turn . 2.1 Unsupervised evaluation In this setting the results of
P11-1148 WSD task , using recall . In the unsupervised evaluation , the induced senses are evaluated
P12-1090 ii ) of several supervised and unsupervised evaluation metrics and ( iii ) of various
S10-1011 section presents the measures of unsupervised evaluation , i.e V-Measure ( Rosenberg and
S07-1002 is not straightfor - ward . The unsupervised evaluation seems to be sensitive to the
hide detail