Skip to main content
Cross Validated

Return to Question

edited title and tags for searchability
Source Link
Andre Silva
Andre Silva
  • 3.1k
  • 5
  • 33
  • 57

How to improve Improving SVM performance on data with missing features and outliers?

I'm trying to learn R for ML purposes, and right now i'mI'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and e1071 SVM. My results are quite good: 80% quality on cross-validation.

Question: Is there any best practices or advices for building classifier on such a bad data? Maybe Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

How to improve SVM performance on data with missing features and outliers?

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and e1071 SVM. My results are quite good: 80% quality on cross-validation.

Question: Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

Improving SVM performance on data with missing features and outliers?

I'm trying to learn R for ML purposes, and right now I'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and e1071 SVM. My results are quite good: 80% quality on cross-validation.

Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

added 98 characters in body; edited tags; edited title
Source Link
chl
chl
  • 54.3k
  • 23
  • 227
  • 388

Improve How to improve SVM performance on data with missing features and outliers in R?

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from AmeliaAmelia package and 'e1071'e1071 SVM. My results are quite good: 80% quality on cross-validation.

Question is:Question: Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

Thank you.

P.S. : Sorry for bad english

Improve SVM performance on data with missing features and outliers in R

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and 'e1071' SVM. My results are quite good: 80% quality on cross-validation.

Question is: Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

Thank you.

P.S. : Sorry for bad english

How to improve SVM performance on data with missing features and outliers?

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and e1071 SVM. My results are quite good: 80% quality on cross-validation.

Question: Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

Source Link
Philipp G. Sinicyn
Philipp G. Sinicyn
  • 131
  • 1
  • 3

Improve SVM performance on data with missing features and outliers in R

I'm trying to learn R for ML purposes, and right now i'm building classifier for my data (10 dimensions, ~400 elements, 2 classes), which have some outliers in it, and a lot of missing values.

I'm using multi-imputation of missing values from Amelia package and 'e1071' SVM. My results are quite good: 80% quality on cross-validation.

Question is: Is there any best practices or advices for building classifier on such a bad data? Maybe I should somehow filter outliers first? Maybe I should consider some other method for imputing missing values?

Thank you.

P.S. : Sorry for bad english