weighted features in SVM classification
3 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
hi dear all I work on weighted classification by SVM in matlab as part of my work , and in my work I want to apply weight of features for every features used as input of svm as an example if data set is input= {1 2 3; 4 5 6; 7 8 9; 10 11 12}. output={-1,+1,-1,+1} I want to apply a weight vector such as {0.5 1 2} so every features in first column of input = 1 4 7 10 have 1/4 importance of features in the third column = 3 6 9 12 in the svm margin computing is there possible by matlab itself? or if not were i can find a good source code of svm written by matlab? tnx very lot
3 comentarios
Konstantinos Papazachariou
el 28 de En. de 2017
Editada: Konstantinos Papazachariou
el 28 de En. de 2017
Hi there! I am interesting to apply weights to my feature vector as well! I am not sure if rankfeatures can help. Anyone knows more on that matter?
Ronghui Yan
el 13 de Sept. de 2017
Hi, i am also working on the feature weighted SVM. Could you please share your source code? Thanks a lot.
Respuestas (0)
Ver también
Categorías
Más información sobre Classification en Help Center y File Exchange.
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!