CODEWITHSUNDEEP
machine-learningdata-miningsvmlibsvmliblinear
chet
chet

Reputation: 627

Instance weighing in Libsvm / Liblinear

I often use the instance weights with Libsvm for classification problems. http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/#weights_for_data_instances

Does anyone know the details of the algorithm that is implemented when one uses instance weighing in libsvm? The standard SVM model learning algorithm assigns equal weights to all training instances, and thus to the error on the training instances. I believe that the algorithm that Libsvm uses would be different. Upon searching online, I do find some papers that do something similar. For example [1] but I need to confirm with someone who may be sure about this.

Thanks!

[1] Yang, Xulei, Qing Song, and Yue Wang. "A weighted support vector machine for data classification." International Journal of Pattern Recognition and Artificial Intelligence 21.05 (2007): 961-976.

Upvotes: 3

Views: 1298

Answers (1)

lejlot
lejlot

Reputation: 66815

There is no "special algorithm", simply, in "equal weight" SVM you have a "C" weight

1/2 ||w||^2 + C SUM_i xi_i

which in case of samples weights s_i simply becomes

1/2 ||w||^2 + C SUM_i s_i xi_i

that's all, it is exactly the same as having different cost cosntant C for each sample

Upvotes: 5

Related Questions