The Altair Community is migrating to a new platform to provide a better experience for you. The RapidMiner Community will merge with the Altair Community at the same time. In preparation for the migration, both communities are on read-only mode from July 15th - July 24th, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here.

Dimensionality reduction using ICA

ThiruThiru Member Posts: 100 Guru
Rapidminer has got an operator 'ind. component analysis' for dimensionality reduction. We have to choose 'fixed no' in parameter configuration of the operator and accordingly the number of attributes are chosen. 

unlike PCA - where the reduction of attribute is done based on decreasing order of eigen values /variance ,

in ICA - it is not clear on what basis the features are chosen.  In my case:  i have six attributes means , i get ic 1, ic2 to ic6. 
Number of Components: 6
Resulting attribute weights (from first component):
attr1: 0.019
attr2: -0.040
attr3: -0.055
attr4: 0.135
attr5: -0.115
attr6: -0.955

 if I choose 4 attribute  it end up in ic1 , ic2 , ic3 ,  ic4.  But it is not clear how these independant components are chosen for reduced dimension requirement.  Please let me know. thanks.


Sign In or Register to comment.