Data in the last column

hi please mention why we have these indices at the end.

Constrained minimization technique for topic identification using discriminative training and support vector machines . latent semantic indexing matrix ; discrimina-tive training ; constrained minimization approach ; support vector machines ; banking call routing ; combination strategy ; classification error ; classification accuracy ; classifier accuracy ; switchboard databases ; vector-space model ; lsi matrix ; baseline classifiers ; score separation ; classifiers ; ensemble ; classifier ; accuracy 17 5 16 ; 1 0 11 ; 2 0 11 ; 2 0 7 ; 1 0 6 ; 16 4 12 ; 17 5 12 ; 3 1 0 ; 14 1 2 this paper describes the to combine multiple in order to improve . since errors of individual in the should somehow be uncorrelated to yield higher , we propose a where the combined is a function of the correlation between classification errors of the individual . to obtain powerful single , different techniques are investigated including and , which is a popular . we also investigate of the on . minimizes the by increasing the of the correct from competing documents . experimental evaluation is carried out on a and on with a set of 23 and 67 topics respectively . results show that the combined we propose outperforms the of individual by 44 % . 2 14 7 22 27 18 -1 15 5 8 18 -1 3 0 10 26 18 -1 1 11 20 21 18 -1 6 13 23 18 -1 18 -1 4 9 19 24 25 18 -1

and what is sorder and whats it use?

rikdz / GraphWriter

Data in the last column #22