Similarity based ML
k-Nearest Neighbor, Support Vector Machines, One-Classification SVM, Logistic Regression
K-Nearest Neighbor
Input Data
*Target Data | Input data |
*Model Name | Name of the model |
Output Name | Name of the output after calculation(optional) |
Merge Data | If not empty, the prediction data will be stored in the selected data frame(optional) |
*X | Input Variable |
*Y(factor) | The target variable for classification or regression. If it is not a factor type, convert it to a factor. |
Support Vector Machines
Input Data
Parameters | Input |
*Target Data | Input data |
*Model Name | Name of the model |
Output Name | Name of the output after calculation(optional) |
Merge Data | If not empty, the prediction data will be stored in the selected data frame(optional) |
*X | Input variable |
*Y | Target data. Factor type for classification and numeric type for regression. |
For classification, select C-classification of nu-classification for the type under Arguments. For regression, select eps-regression or nu-regression.
One-Classification SVM
Input Data
X | Input variables |
One-classification SVM is used for the novelty detection. It uses the method 'one-classification' in the svm function.
Logistic Regression
Input Data
X | Independent variables |
Y | Dependent categorical variable |
Workflow Example
R-Flow Task Example Video: Support Vector Machines
R Packages
K-Nearest Neighbor
Packages name: kknn, Method: kknn
https://cran.r-project.org/web/packages/kknn/index.html
Support Vector Machines/One-Classification svm
Packages name: e1071, Method: svm
Last updated
